Short Synthesis of Structurally Diverse N-Acylhomoserine Lactone Analogs and Discovery of Novel Quorum Quenchers Against Gram-Negative Pathogens

Abstract

Quorum quenchers are emerging as an alternative to conventional antimicrobials, since they hinder the development of virulence or resistance mechanisms but without killing the microorganisms, thus, reducing the risk of antimicrobial resistance. Many quorum quenchers are analogs of the natural quorum-sensing signaling molecules or autoinducers. Thus, different analogs of natural N-acylhomoserine lactones (AHLs) have been reported for controlling virulence or reducing the production of biofilms in Gram-negative pathogens. Herein we report the preparation of AHL analogs with a variety of N-substituents in just two steps from readily available N-substituted hydroxyproline esters. The substrates underwent an oxidative radical scission of the pyrrolidine ring. The resulting N-substituted β-aminoaldehyde underwent reduction and in situ cyclization to give a variety of homoserine lactones, with N- and N,N-substituted amino derivatives and with high optical purity. The libraries were screened for the inhibition of violacein production in Chromobacterium violaceum, a Gram-negative pathogen. For the first time, N,N-disubstituted AHL analogs were studied. Several N-sulfonyl derivatives, one carbamoyl, and one N-alkyl-N-sulfonyl homoserine lactone displayed a promising inhibitory activity. Moreover, they did not display microbicide action against S. aureus, C. jejuni, S. enterica, P. aeruginosa, and C. albicans, confirming a pure QQ activity. The determination of structure–activity relationships and in silico ADME studies are also reported, which are valuable for the design of next generations QQ agents.

1. Introduction

Antimicrobial resistance is a major threat to health and food security, according to the WHO and FAO, and it is urgent to develop new treatments [1,2]. Among the alternatives to current antimicrobials, one of the most promising is the discovery of quorum quenchers (QQs), compounds that disrupt the microbial communication systems known as quorum sensing (QS) [3,4,5]. The latter mediates coordinated actions, such as the activation of virulence and coordinated attacks on the host, production of toxins and proteolytic enzymes, swarming, generation of defensive biofilms, and activation of other resistance mechanisms [3,4,5,6,7,8,9,10]. Quorum sensing has been discovered both in bacteria [4,6,7,8] and fungi [9,10,11,12], and there is also an interkingdom communication that modulates the host defensive responses [13,14,15,16,17,18]. Therefore, the molecules that are able to regulate QS could be critical to prevent or reduce pathogenicity [3,4,5,19,20,21,22,23,24,25]. In addition, since pure QQs do not display a microbicidal action, the pressure towards the emergence of resistance is greatly reduced [3,4,5,26].

There are many QS systems, but all are mediated by signaling molecules known as autoinducers [3,4,5,6,7,8,9,10,27]. Among the most important QS signals are N-acylhomoserine lactones (AHLs), which are produced and/or recognized by Gram-negative bacteria [17,18], including pathogens such as Escherichia coli., Salmonella sp., or Pseudomonas aeruginosa [6,27,28], and they are also involved in interkingdom communication [17]. For instance, homoserine lactone (1, 3OC8CHSL, Figure 1) is a quorum-sensing signal in the biofilm-producing pathogen Pseudomonas aeruginosa [6], which causes severe complications in cystic fibrosis and contaminates burns, wounds, and even medical devices.

Due to their importance, different AHL analogs have been developed as potential QQs, either by replacing the lactone ring by other hetero- or carbocycles and even a few acyclic chains [29,30,31,32,33,34,35], or by changing the N-substituents [36,37,38,39]. In other cases, the natural acyl chains have been replaced by synthetic acyl and 3-oxoacyl chains with unnatural carbocyclic, aromatic, or heteroaromatic groups and also with heteroatoms (e.g., N, S, O) inserted in the alkyl chains [6,36,37,38,39,40,41,42,43,44,45,46]. For instance, Bassler et al. reported that synthetic compound (2), mBTL, an analog of compound (1) containing a thiolactone ring and a modified acyl chain, inhibited the generation of biofilms and of the virulence factor pyocianin (IC₅₀ = 4 µM). Moreover, mBTL protected human epithelial cells and even pluricelullar organisms such as the nematode Caenorhabditis elegans from death caused by P. aeruginosa [6].

The N-acyl group has also been replaced by thiocarbamoyl and carbamoyl groups [47,48], sulfonamides [49,50,51], and sulfonylureas [49,52]. For instance, Queneau and Soulére reported that p-nitrobenzylcarbamate 3a and its thiocarbamate analog 3b (Figure 1) had a promising quorum-quenching activity in Vibrio fischeri. Both compounds were similarly active with an IC₅₀ value of about 20 µM [47]. In another example, different compounds were tested as competitive inhibitors of 3-oxohexanoyl-L-homoserine lactone, the ligand of transcriptional regulator LuxR in Vibrio fischeri. The most active compounds inhibited bioluminescence in the pathogen. Amide inhibitor 4a was taken as reference (IC₅₀ = 2 µM). When the amide function was replaced by a sulfonamide (compound 4b), the activity was considerably reduced. However, when the chain length was slightly decreased (compound 4c), the activity was similar to the reference compound. Moreover, replacement of the previous chain by a pentenyl group (compound 4d) also gave good results. Increasing the acyclic chain length (as in compound 4e) again decreased the inhibitory activity [50]. This example shows how structural fine-tuning can have an important impact in bioactivity.

Blackwell [30,34,39,40,46], Nagarajan [34], Subba Reddy and Padmajan [49], and other groups have reported important collections of potential quorum-sensing inhibitors. Some of these replacements have yielded potent quorum quenchers that are being evaluated for medical use. These encouraging results have fueled more work in the area. We noticed, for instance, that most sulfonamides are alkylsulfonyl derivatives [49,50], while there are few examples of N-arylsulfonyl AHL, mostly p-aminophenylsulfonyl derivatives [49,51]. To our surprise, the substituted benzamido groups are also scarcely studied [36,37,38,39,45,46], the same as carbamoyl substituents [47]. Moreover, there were no reports on the activity of N,N-disubstituted AHLs. In order to study these new derivatives, a versatile synthetic methodology was optimized, starting from low-cost substrates derived from natural 4-hydroxy-L-proline (Hyp, Figure 2). This methodology avoids commercial amino-γ-lactone as a substrate, which is prone to epimerization during the N-acylation/functionalization step. Instead, the hydroxyproline derivatives 5 would undergo an oxidative radical fragmentation of the pyrrolidine ring through the C₄-C₅ bond, to give aldehydes 6. This scission was previously reported for N-carbamoyl and N-acyl derivatives but not for sulfonamides [53,54,55,56,57]. Then, the aldehydes 6 would be transformed into the lactones 7 using a reduction–lactonization reaction. It must be noted that the method reported herein affords both N-substituted (X = O, R = H) and N,N-disubstituted (R, Z ≠ H) amino lactones 7 in a simple way. Moreover, small variations in the proposed synthetic route could afford different heterocycles (e.g., reductive aminations would yield X = N-alkyl), although the present work is devoted to the lactones.

2. Results and Discussion

The subheadings of this section describe experimental results and their interpretation.

2.1. Preparation of Libraries

The libraries were prepared from a variety of N-substituted hydroxyproline substrates 5a–p (

Table 1. Synthesis of aldehydes 6a–p by oxidative radical scission of the hydroxypyrrolidines 5a–p.

entry	substrate	Z	Product (%)
1	5a	X = H	6a (80)
2	5b	X = Me	6b (73)
3	5c	X = Cl	6c (73)
4	5d	X = I	6d (86)
5	5e	X = NO2	6e (71)
6	5f	X = H	6f (88)
7	5g	X = F	6g (85)
8	5h	X = Cl	6h (82)
9	5i	X = I	6i (73)
10	5j	X = NO2	6j (87)
11	5k		6k (78)
12	5l	Ac	6l (72)
13	5m		6m (70)
14	5n	Boc	6n (70)
15	5o	Cbz	6o (56)
16	5p	CO2Ph	6p (79)

). Their conversion into the aldehydes 6a–p has been developed by our group [53,54,55,56,57]. Under treatment with (diacetoxyiodo)benzene (DIB) and iodine, a 4-hypoiodite is formed. Then, irradiation with visible light provides energy for the homolytic cleavage of the O-I bond. The resultant O-radical undergoes a regioselective β-fragmentation, and, thus, the C₄-C₅ bond is cleaved and a N-methyl radical is formed. This radical species is stabilized by the nitrogen function, which accounts for the regioselectivity observed. However, it quickly reacts with iodine generating an unstable N-CH₂-I moiety. Extrusion of iodide gives intermediate iminium ions 8a–p, which are trapped by acetate ions from the DIB reagent, yielding the aldehydes 6a–p [53,54,55,56,57]. The process took place in good yields, and the resulting products presented an α-chain and an N,O-acetal, which could be manipulated independently, as shown below.

Interestingly, this is the first comparison of an oxidative radical scission generating an N-sulfonyliminium ion and related scissions affording an N-acyliminium ion intermediate. The sulfonamides 5a–e gave fragmentation yields similar to the benzamides 5f–3j. The acyl groups with alkyl chains and the carbamoyl groups also gave good, although slightly lower, yields. It should be noted that the scission of substrate 5m (entry 13) gave only one enantiomer, showing that the scission proceeded without epimerization.

The reduction of the aldehydes and in situ lactonization was carried out under different conditions. Using the standard conditions of sodium borohydride in methanol, the reduction of the aldehyde was accompanied by cleavage of the acetoxymethyl group, to give lactones 7a–p (

Table 2. Synthesis of lactones 7a–p by reduction–cyclization of aldehydes 6a–p.

entry	substrate	Z	Product (%)
1	6a	X = H	7a (63)
2	6b	X = Me	7b (65)
3	6c	X = Cl	7c (29)
4	6d	X = I	7d (52)
5	6e	X = NO2	7e (56)
6	6f	X = H	7f (67)
7	6g	X = F	7g (67)
8	6h	X = Cl	7h (40)
9	6i	X = I	7i (59)
10	6j	X = NO2	7j (53)
11	6k		7k (34)
12	6l	Ac	7l (43)
13	6m		7m (57)
14	6n	Boc	7n (70)
15	6o	Cbz	7o (30)
16	6p		7p (23)

). In effect, traces of sodium methoxide formed in situ caused the saponification of the acetate, and the lactonization released more methoxide. Under these conditions, a small epimerization was detected, as evidenced by changes in the optical rotation of the product in processes carried out at different reaction times. In order to avoid it, an optimized process was developed. Thus, after the reduction a quick work-up was performed, the solvent was removed, and the residue was dissolved in dichloromethane, treated with triethylamine and refluxed for 1 h. Under these mild conditions, lactones 7a–p were obtained as shown in Table 2, with reproducible optical rotations, and moreover, the lactonization of substrate 6m provided a single isomer. The cyclization of the phenyl carbamate 6p deserves comment, as the cyclization proceeded to the six-membered carbamate 7p and not to the desired five-membered lactone, thus, demonstrating that the phenoxy function is an excellent leaving group.

The preparation of N,N-disubstituted homoserine lactones required a different procedure, using our reported conditions for the reduction of N-acetoxymethyl groups [56]. The selected aldehyde substrates were the sulfonamide 6b, the benzamide 6g, and the carbamate 6p (Figure 3), as representative examples of the most frequent protecting groups, Z. Therefore, these substrates were treated with triethylsilane in the presence of boron trifluoride etherate. In the non-polar solvent, the N,O-acetal would generate an iminium ion, which would be reduced by the silane to a N-methyl group.

The reduction of sulfone 6b afforded the N-methyl lactone 7q in 62% yield. To our surprise, in the case of benzamide substrate 6g the N-methyl ester was obtained as the minor product (31%), the major being the N,O-acetal 7s (60%). The carbamate substrate 6p yielded a mixture of the N-methyl 7t and N-acetoxymethyl 7u products in a 1:1 ratio (98% yield).

These results point to a probable mechanism of the process. As shown for compounds 7t/1u, the lactone is formed first to give compound 7u, and then the acetoxymethyl group is transformed into an imine, which is reduced to an N-methyl group, affording compound 7t. Since the conversion of 7u into 7t was incomplete, a mixture of the two compounds was isolated. In the case of the sulfonamide substrate 6b, the reaction was completed to give only compound 7q. The sulfonyl protecting group makes the imine intermediate more electrophilic than the carbamate-protected imine and, therefore, more reactive with the silane. In the case of benzamides 7r/7s, an intermediate N-acetoxymethyl lactone similar to 7u is likely formed. Then, the N-acetoxymethyl group evolves to an imine, which is either reduced to an N-methyl group (compound 7r) or trapped by methoxy ions to give the methoxy derivative 7s. The methoxy ions are formed from the methyl ester during the intramolecular lactonization reaction.

The introduction of R = alkyl likely alters interactions with the biological target with respect to N-monosubstituted AHLs (R = H), allowing interesting structure–activity relationships to be determined.

In summary, a library of AHLs with a variety of N-substituents and a library of AHL aldehyde precursors were prepared in good yields and from readily available, low-cost hydroxyproline substrates. The evaluation of their quorum-quenching and antimicrobial activities is detailed below.

2.2. Evaluation of Quorum Quencher and Antimicrobial Activities

The purpose of the libraries is to identify a compound with quorum-quenching activity but not bactericidal action, so that it can prevent bacterial infections but avoid damage to these microorganisms. In this way, the risk of eliciting antibiotic resistance is greatly reduced and the beneficial microbiota is spared. Therefore, the best quorum quenchers should display negligible antibiotic action.

To determine the quorum-quenching activity, the reporter strain Chromobacterium violaceum CECT 494 (also called ATCC 12472) was used [18,51,58,59]. This Gram-negative pathogen has a LuxIR-type circuit, called CviIR, which regulates the production of the autoinducer N-decanoyl homoserine lactone (C10-HSL) [58,59]. When the autoinducer released into the extracellular space reaches a certain threshold, it re-enters the cytoplasm and binds to the transcriptional activator CviR, activating the expression of genes necessary for the production of the pigment violacein [13,58,59,60]. Therefore, treatment of the CECT 494 strain with quorum quenchers will decrease the generation of the violet pigment, which could be measured by a colorimetric assay (Figure 4), according to the procedure reported by Choo et al. [60].

The results are shown in

Table 3. Results of the IQS detection assay in the C. violaceum CECT 494 strain.

Compound	C (μM)	Inhibition of Violacein Production (%) μ ± DE a
7a	200	70.42 ± 3.55
7a	100	57.27 ± 5.40
7a	50	41.94 ± 4.64
7b	200	62.10 ± 3.89
7b	100	49.59 ± 6.16
7b	50	31.73 ± 7.31
7c	200	51.55 ± 3.62
7c	100	44.39 ± 7.43
7c	50	39.42 ± 4.90
7d	200	26.37 ± 10.15
7e	200	48.74 ± 5.69
7e	100	36.96 ± 7.51
7f	200	4.16 ± 3.41/NS
7g	200	26.21 ± 7.63
7h	200	32.80 ± 3.38
7i	200	23.34 ± 4.73
7j	200	NI
7k	200	6.98 ± 3.02
7l	200	NI
7m	200	NI
7n	200	30.47 ± 1.82
7o	200	67.06 ± 3.82
7o	100	51.94 ± 2.21
7o	50	43.76 ± 1.28
7p	NT	NT
7q	200	43.52 ± 4.72
7r	200	11.79 ± 3.77/NS
7s	200	8.19 ± 1.47/NS
7t	200	7.36 ± 2.25/NS
7u	200	10.87 ± 3.72/NS

^a The results are shown as percentages of inhibition of violacein production by treatment with the α-amino-γ-lactone derivatives, compared to an untreated control. The results are given as the average percentage of inhibition ± standard deviation (n = 3). The values show significant differences (p < 0.05) with respect to the non-treated control according to the one-way ANOVA statistical procedure. NI: non-inhibition. NS = no significant differences with untreated control. NT = Not tested. The most relevants results are in bold, and for inhibition >50% are highlighted in red.

, with respect to an untreated control. In addition, since the benzylcarbamate 7o is a known quorum quencher, it was used as a positive control [47] while the known inactive N-acetyl derivative 7l was used as a negative control [61]. It should be said that although 7l possesses the N-acyl homoserine lactone moiety, the size of the N-acyl chain is small, causing the loss of activity. In contrast, many homoserine lactones with bulkier N-substituents (such as 1o) are usually active for a variety of Gram-negative bacteria, such as Chromobacterium violaceum, Vibrio fischeri, Escherichia coli, etc.

The most active compounds were the sulfonamides 7a–c and 7e, the reference compound benzyl carbamate 7o, and the N-methyl toluenesulfonamide 7q. It was ruled out that the reduction in pigment production was due to growth inhibition, since the count of viable colony-forming units gave similar values for the treated biosensor and the untreated control.

The most potent sulfonamides were studied at three doses of 200, 100, and 50 µM. Compound 7a achieved about 70% inhibition of violacein production at 200 µM, but even when the dose was successively halved, the inhibition did not decrease proportionally but was maintained at satisfactory levels (57% and 42% for 100 and 50 µM). Compound 7b, which displayed 62% inhibition at 200 µM, also maintained a good activity when the dose was reduced (50% and 32% for 100 and 50 µM). Interestingly, the halo derivatives did not increase inhibition, although the p-chlorophenylsulfonamide 7c also displayed a satisfactory activity at 200 µM (51% inhibition), which was only slightly reduced at 100 µM (44%). The p-iodophenylsulfonamide 7d had a much lower activity, perhaps because the steric hindrance of the iodo group complicated the interaction with the Lux receptor. In fact, comparing 7a and 7b, it is clear that the p-alkyl substituent led to lower activity. When the p-substituent was the nitro group (compound 7e), the activity was somewhat recovered (49% at 200 µM but 37% at 100 µM).

In contrast to the sulfonamides, the benzamides 7f–m displayed little quorum-quenching activity. However, the reference benzyl carbamate 7o displayed the second best inhibition (67%, 52%, and 44% at 200, 100, and 50 µM, respectively). The activity dropped when the bulky t-butyl carbamate 7n was used (30% at 200 µM).

With respect to the role of N-substitution, when toluenesulfonamide 7b was compared with its N-methyl analog 7q, it was observed that higher substitution decreased activity (62% for 1b and 44% for 1s at 200 µM). The same happened when the benzamide 7g was compared with the N-methyl analog 7r (26% vs. 12% inhibition at 200 µM). When the N-methyl group was replaced by a bulkier N-methoxymethyl moiety, the activity was further reduced (about 8% inhibition at 200 µM). The substituted phenyl carbamoyl derivatives 7t and 7u displayed little quorum-quenching activity, supporting that N-substitution is deleterious for quorum quenching.

A summary of the dose–effect relationship for the most promising compounds (sulfonamides 7a–c and 7e, benzyl carbamate 7o, and the N-methyl toluenesulfonamide 7q) is shown in Figure 5. As commented before, the inhibitory effect slowly decreases upon lowering the dose, but a satisfactory activity is, nevertheless, maintained.

As commented on in the introduction, different groups have worked on the development of quorum-sensing modulators, and many inhibitors have been discovered. However, active work takes place in the area. Most of the work on sulfonamides has been carried out with alkylsulfonyl derivatives [49,50], as shown by compounds 4b–4e in Figure 1 [50], where small changes in the chain length and substituents can notably alter the quorum-quenching activity. In contrast, there are few examples of the N-arylsulfonyl AHL [49,51], mostly p-amidophenylsulfonyl derivatives with bulky N-acyl substituents [51]. However, Reddy and Padmaja report the p-toluene sulfonamide 7b and the p-nitrophenyl derivative 7e, as well as the p-amino analog of 7e [49]. The halogenated derivatives 7c–d and the unsubstituted compound 7a were not tested. The only compound with significative QSI activity was 7b, which matches our results where 7b had a considerable quorum-quenching activity, quite superior to 7e. However, our results show that a simpler sulfonamide 7a is the most potent derivative, and more importantly, that a considerable activity is retained when the dose is reduced. As commented on later, there are other advantages with respect to 7b: the lack of antimicrobial activity for the tested Gram-negative and Gram-positive pathogens and a low risk of eliciting antimicrobial resistance.

For the first time, very related sulfonamides and benzamides are compared, with the first displaying a promising activity, in contrast with the second, whose inhibitory activity was quite low. It is interesting that while the literature reports many examples of N-acyl homoserine lactones, including examples where alkyl chains are attached to aromatic groups [36,37,38], the N-benzoyl derivatives are scarcely reported [39,45,46]. Blackwell et al. carried out the most complete study for benzamides [39,46], with a p-bromo benzamide being a potent QscR antagonists in P. aeruginosa [46]. However, when a set of benzamides was studied as potential quorum-sensing modulators in E. coli, the benzamides were among the few compounds that did not activate the promiscuous SdiA receptor [39]. In our case, the benzamides had little activity on the CviR receptor in C. violaceum, even the p-chloro and p-iodo derivatives 7h and 7i. Therefore, the sulfonamide derivatives are preferred to the benzamides.

The carbamoyl and thiocarbamoyl substituents have received some attention, as commented on in the Introduction for Vibrio fischeri [47]. Interestingly, both types of carbamate displayed similar activity in the reported examples by Queneau et al. The benzyl carbamates (and particularly the p-nitrobenzyl derivatives) were the most potent [47], as in our case, while the t-butyl carbamate showed a relatively small activity.

Finally, for the first time, this article compares the activity of N-methyl derivatives with the demethylated products (7q vs. 7b), showing that the second gave superior inhibition. These results support that the binding of the compounds to the quorum-sensing receptor CviR is enhanced by a hydrogen bond between the N-H group and the receptor. N,N-disubstituted AHL analogs would lack the ability to form this bond, and the interaction would decrease.

The antibiotic activity was then checked. As commented before, it was observed that the lactones did not affect bacterial growth and the number of colony-forming units of C. violaceum. However, they could have antimicrobial activity against other microorganisms, in particular the sulfonamide derivatives. Therefore, the broth microdilution method [62,63] was used to identify the compounds that at 200 μM presented activity against the Gram-positive pathogen Staphylococcus aureus CECT 794 and the Gram-negative bacteria Campylobacter jejuni CECT 9112, Salmonella enterica CECT 456, and Pseudomonas aeruginosa CECT108 (

Table 4. Antimicrobial activity of aldehydes 6a–6p against bacterial pathogens using the disk diffusion assay.

Compound	Zone of Inhibition (mm)
	S. aureus (CECT 794)	C. jejuni (CECT9112)	S. enterica (CECT456)	P. aeruginosa (CECT108)
6a	------	------	------	------
6b	13.30 ± 0.60	15.00 ± 1.20	NI	NI
6c	13.70 ± 0.30	14.00 ± 0.00	NI	NI
6d	12.00 ± 1.00	12.00 ± 1.20	NI	NI
6e	12.30 ± 0.30	17.00 ± 0.60	NI	NI
6f	NI	NI	NI	NI
6g	NI	NI	NI	NI
6h	NI	NI	NI	NI
6i	NI	NI	NI	NI
6j	NI	NI	NI	NI
6k	13.00 ± 0.90	16.00 ± 0.60	NI	NI
6l	NI	NI	NI	NI
6m	NI	NI	NI	NI
6n	NI	NI	NI	NI
6o	NI	NI	NI	NI
6p	NI	NI	NI	NI
Tetracycline	26.00 ± 1.00	25.00 ± 1.20	24.00 ± 1.60	15.00 ± 2.90

Results as average of inhibition zone ± standard deviation (n = 3) in millimeters (mm). NI: No inhibition.

). None of the compounds displayed a minimum bactericidal concentration (MBC) or minimum inhibitory concentration (MIC) below 200 μM against S. enterica and P. aeruginosa. The p-nitrophenylsulfonamide lactone 7e presented an MIC 101–150 μM against S. aureus, and the dinitrobenzamide compound 7k presented an MIC in the range 155–199 μM against S. aureus and C. jejuni. This low antimicrobial activity is a requisite for pure QQ agents. To our satisfaction, the most active QQs 7a–c, 7e, 7o, and 7q had negligible direct antimicrobial action.

The aldehyde precursors 6b–p were also tested using EUCAST strains [64,65] as shown in Table 4. All the sulfonamides 6b–e displayed a promising activity against S. aureus and C. jejuni, although still inferior to the antibiotic standard (tetracycline). The dinitrobenzamide 6k was also active against S. aureus and C. jejuni. None of the aldehydes displayed activity against S. enterica and P. aeruginosa.

The contrast between the antimicrobial activity of the lactones 7b–e and the aldehydes 6b–e suggest that the 4-carbonyl group in the latter interacts with nucleophilic moieties in the receptors. The study of the antimicrobial activity of these aldehydes is in course and will be published in due time.

2.3. In Silico ADME Study of the AHL Analogs

In order to determine whether the lactones and the most active aldehydes had appropriate ADME properties, an in silico study was carried out using the SwissADME tool (www.swissadme.ch), accessed on 8 February 2025 [66,67]. The results are shown in

Table 5. Summary of in silico physicochemical properties for lactones 7a–o and 7q–u and selected aldehydes 6a–e and 6k.

Compound	MW (g/mol)	N° H-Bond Donors	N° H-Bond Acceptors	N° Rotable Bonds	TPSA (Ų)	LogPo/w	LogS (SILICOS-IT)
7a	241.26	1	5	3	80.85	0.90	−2.89 Soluble
7b	255.29	1	5	3	80.85	1.23	−3.27 Soluble
7c	275.71	1	5	3	80.85	1.41	−3.50 Soluble
7d	367.16	1	5	3	80.85	1.53	−3.78 Soluble
7e	286.26	1	7	4	126.67	0.22	−2.74 Soluble
7f	205.21	1	3	3	55.40	1.26	−2.91 Soluble
7g	223.20	1	4	3	55.40	1.58	−3.19 Soluble
7h	239.65	1	3	3	55.40	1.80	−3.53 Soluble
7i	331.11	1	3	3	55.40	1.94	−3.83 Soluble
7j	250.21	1	5	4	101.22	0.70	−2.77 Soluble
7k	295.21	1	7	5	147.04	0.07	−2.62 Soluble
7l	143.14	1	3	2	55.40	−0.05	−0.75 Soluble
7m	382.41	2	5	10	93.73	2.18	−5.96 Mod. Sol.
7n	201.22	1	4	4	64.63	0.95	−1.34 Soluble
7o	235.24	1	4	5	64.63	1.42	−3.06 Soluble
7p	159.14	1	4	2	64.63	0.01	−0.50 Soluble
7q	269.32	0	5	3	72.06	1.45	−2.94 Soluble
7r	237.23	0	4	3	46.61	1.75	−2.86 Soluble
7s	267.25	0	5	5	55.84	1.74	−3.00 Soluble
7t	235.24	0	4	4	55.84	1.55	−2.33 Soluble
7u	293.27	0	6	7	82.14	1.54	−2.40 Soluble
6a	343.35	0	8	10	115.43	0.77	−2.58 Soluble
6b	357.38	0	8	10	115.43	1.11	−2.96 Soluble
6c	377.80	0	8	10	115.43	1.32	−3.17 Soluble
6d	469.25	0	8	10	115.43	1.28	−3.42 Soluble
6e	388.35	0	10	11	161.25	0.22	−2.41 Soluble
6k	397.29	0	10	12	181.62	−0.04	−2.29 Soluble

and

Table 6. Summary of in silico pharmacological properties for lactones 7a–p and 7q–u and selected aldehydes 6a–e and 6k.

Compound	GI Absorp	BBB Permeant	P-gp Substrate	CYP Inhibitor	Log Kp (cm/s)	Druglikeness: Lipinki, Ghose, etc.	Abbot Bio. Score	PAINS/Brenk Alerts
7a	High	No	No	No	−7.08	Yes, 0 violations	0.55	0 alerts
7b	High	No	No	No	−6.91	Yes, 0 violations	0.55	0 alerts
7c	High	No	No	No	−6.85	Yes, 0 violations	0.55	0 alerts
7d	High	No	No	No except CYP2C19	−7.39	Yes, 0 violations	0.55	P: 0 alerts; B: Iodo
7e	High	No	No	No	−7.48	Yes, 0 violations	0.55	P: 0 alerts; B: Nitro
7f	High	Yes	No	No	−6.64	Yes, 0 violations	0.55	0 alerts
7g	High	Yes	No	No	−6.67	Yes, 0 violations	0.55	0 alerts
7h	High	Yes	No	No except P450 1A2	−6.41	Yes, 0 violations	0.55	0 alerts
7i	High	Yes	No	No except CYP1A2	−6.94	Yes, 0 violations	0.55	P:0 alerts; B: Iodine
7j	High	No	No	No	−7.04	Yes, 0 violations	0.55	P:0 alerts; B: Nitro
7k	Low	No	No	No	−7.43	No, Veber rules (TPSA > 140) and Egan: TPSA > 131	0.55	P:0 alerts; B: Nitro
7l	High	No	Yes	No	−7.44	No, Muegge and Ghose, low MW	0.55	0 alerts
7m	High	No	Yes	No except CYP3A4	−6.60	Yes, 0 violations	0.55	P:0 alerts; B: >2 esters
7n	High	No	No	No	−6.80	Yes, 0 violations	0.55	P: 0 alerts; B: >2 esters
7o	High	Yes	No	No	−6.65	Yes, 0 violations	0.55	P: 0 alerts; B: >2 esters
7p	High	No	No	No	−7.25	No, Muegge and Ghose, low MW	0.55	P: 0 alerts; B: >2 esters
7q	High	Yes	No	No except CYP2C19	−6.86	Yes, 0 violations	0.56	0 alerts
7r	High	Yes	No	No	−6.63	Yes, 0 violations	0.55	0 alerts
7s	High	Yes	No	No	−6.85	Yes, 0 violations	0.55	0 alerts
7t	High	Yes	No	No	−6.48	Yes, 0 violations	0.55	0 alerts
7u	High	No	No	No	−6.84	Yes, 0 violations	0.55	0 alerts
6a	High	No	No	No except CYP2C19	−8.07	Yes, 0 violations	0.55	P: 0 alerts; B: aldehyde
6b	High	No	No	Same as 6a	−7.90	Yes, 0 violations	0.55	Same as 6a
6c	High	No	No	Same as 6a	−7.84	Yes, 0 violations	0.55	Same as 6a
6d	High	No	No	Same as 6a	−8.38	Yes, 0 violations	0.55	Same as 6a
6e	Low	No	Yes	No	−8.47	No, high TPSA and rotors, O > 10	0.55	P: 0 alerts; B: aldehyde, NO2
6k	Low	No	Yes	No	−8.43	Same as 6e	0.55	Same as 6e

. Table 5 is devoted to the molecular and physicochemical descriptors, such as the MW, number of rotable bonds, H-acceptors, and H-donors, as well as the Topological Polar Surface (TPSA) [68], a useful descriptor to estimate properties such as absorption, brain access, etc., as commented on later.

Another important parameter is log P_o/w, the partition coefficient of the compound in its neutral form between water and n-octanol, which is critical for barrier crossing and biodistribution. Since SwissADME provides different values obtained from different calculation methods (iLOGP, XLOGP3, WLOGP, MLOGP, and SILICOS-IT), an average “consensus” value is shown in the table [69,70,71,72]. Most of the compounds have a positive Log P and are, therefore, lipophilic. Only the dinitrobenzamide 7l and the acetamide 6k have negative LogP and, thus, have more hydrophilic characteristics.

The solubility can also be calculated using different methods, but the table shows LogS obtained with SILICOS-IT, a fragmental method, which has a high linear correlation between theoretical and experimental values (R² = 0.75) [73]. Most compounds have calculated logS values between 0 and −4 and, therefore, should be soluble or very soluble, and only the dipeptide 7m presents a value between −4 and −6 (moderately soluble).

Table 6 displays the estimated pharmacokinetic parameters/properties and the druglikeness. The gastrointestinal absorption for most compounds was estimated to be high (white in the BOILED-Egg method) [67], except for the p-nitrosulfonamide aldehyde 6e and the dinitroderivatives 7k and 6k. As for the ability to cross the Blood–Brain Barrier (BBB), most compounds could not cross it, with the exception (yolk of the BOILED-Egg method) of unsubstituted or halo-substituted benzamides 7f–i, the benzylcarbamate derivative 7o, and the N-alkyl derivatives 7q–7t, which could be very interesting to fight bacterial infections causing meningitis.

SwissADME also predicts whether a compound can be a substrate for the permeability glycoprotein (P-gp, an ABC transporter) and, therefore, undergo active efflux through membranes, such as the BBB or the GI wall to the lumen [74]. Most of the compounds are not expected to be substrates, with the exception of the acetamide 6l, the dipeptide 6m, and the aldehydes 6e–2k. However, lipophilic compounds may cross barriers in a passive way, as commented before for BBB-permeable compounds.

The interaction of the compounds with cytochromes (CYP) is key for their metabolic transformation and subsequent elimination [75]. There are different CYP isoforms such as CYP1A2, CYP2C19, CYP2C9, CYP2D6, and CYP3A4. Their inhibition promotes unwanted drug–drug interactions, accumulation of the drug or its metabolites, and related side-effects. Fortunately, most AHLs were predicted not to inhibit any of the major isoforms, with the exception of halo derivatives 7d, 7h, and 7i, the dipeptide 7m, and the N-methyl derivative 7q, but in those cases, only one isoform was expected to be affected.

Skin permeation (in cm/s, topical use) was also calculated. When the log Kp exceeds −2.5 cm/s, the molecule presents low skin permeation, which is the case for the studied AHLs and aldehydes. This could be positive for fighting topical infections without causing systemic/intradermal effects [76].

Druglikeness is a qualitative assessment of the oral bioavailability of a drug candidate. Five filters were applied: the Lipinski (Pfizer) [72], Ghose (Amgen) [77], Veber (Glaxo-Smith-Kline) [78], Egan (Pharmacia) [79], and Muegge (Bayer) [80]. The Lipinski and Veber filters are the best known, and the first was implemented as follows: MW < 500, MLOGP < 415, N or O < 10, NH or OH < 5. In addition, the Veber filter requires that n° rotational bonds < 10 and TPSA < 140. In the Ghose filter, 160 < MW < 480 and −0.4 < WLOGP < 5.6, and the number of atoms should be in the range 20–70. The Muegge filter requires that 200 < MW < 600, the number of H-bond acceptors < 10, the number of rotable bonds < 15, TPSA < 150, and −2 < XLOGP < 5. Finally, the Egan filter determines that WLOGP < 5.88 and TPSA < 131.6.

All the compounds met Lipinski’s rules, and only the dinitrobenzamide compound 7k did not meet the Veber and Egan rules due to the high TPSA value. Finally, some compounds (7l, 7p) had a low MW for the Muegge and Ghose rules, and aldehydes 6e and 6k did not meet several criteria, such as suitable TPSA range, number of heteroatoms and rotors, etc. In general, most compounds showed good druglikeness.

The Abbot Bioavailability Score is used to predict the probability of presenting at least 10% oral bioavailability in rat or Caco-2 permeability [81]. Since all the compounds had a 0.55 or higher score, this oral bioavailability criteria was met.

The last column is devoted to PAINS [82] and Brenk [83] alerts. PAINS (pan assay interference compounds) refer to promiscuous compounds that give strong responses in assays for a variety of protein targets (and, therefore, false positives). To our satisfaction, no PAINS alerts were obtained.

The Brenk alarm points out chemical moieties that are known to cause toxicity or instability or those that are dyes. The alarm was obtained for nitrobenzene derivatives (as potential carcinogens), aldehydes (reactive electrophilic moiety), or iodo aromatic groups (depending on the dose may interfere with thyroid function). Since the ester groups may be hydrolyzed by proteases, reducing in vivo stability, some compounds with more than 2 ester groups elicited alarms. It must be said, however, that the Brenk alerts are orientative and do not exclude these compounds from pharmaceutical development; they simply recommend further toxicology or stability assays at early stages. Fortunately, many of our active compounds showed no alerts at all.

Finally, Figure 6 shows the Bioavailability Radar of selected compounds, namely a representation of the oral bioavailability based on their molecular and physicochemical properties. The compounds with predicted good oral availability should fall in the pink area and, therefore, would have MW between 150 and 500 g/mol, TPSA in the 20–130 Ų range, logS < 6, a maximum of 9(10) rotatable bonds for optimum flexibility, and suitable lipophilicity (XLOGP3 between −0.7 and + 5.0). The sulfonamide and benzamide AHL derivatives met this requirement, except for dinitro compound 7k, which was slightly more polar than recommended. The benzyl carbamate 7o and the N-alkyl derivatives 7q–u also met the criteria. However, aldehydes were predicted to not be orally available, due to their high flexibility, and in some cases (6e, 6k) also because they were too polar.

3. Materials and Methods

3.1. Synthetic Procedures and Characterization Data

General Methods. Commercially available reagents and solvents were analytical grade or were purified by standard procedures prior to use. All reactions involving air- or moisture-sensitive materials were carried out under a nitrogen atmosphere. Melting points were determined with a hot-stage apparatus and are uncorrected. Optical rotations were measured at the sodium line at ambient temperature (26 °C) in CHCl₃ solutions. NMR spectra were determined at 500 or 400 MHz for ¹H and 125.7 or 100.6 MHz for ¹³C, at 25 °C or 70 °C, as stated for each case. Sometimes, due to slower rotamer interconversion at 26 °C, two (or more) sets of signals are visible at room temperature, while only one set of signals (rotamer average) is seen at 70 °C, due to faster rotamer interconversion. For some compounds, the ¹H NMR spectra shows some signals as broad bands (br b) due to equilibria between rotamers.

¹H NMR spectra are reported as follows: s = singlet, d = doublet, t = triplet, dd = doublet of doublets, ddd = doublet of doublet of doublets, q = quartet, m = multiplet, br = broad, br b = broad band, and br s = broad singlet; coupling constant(s) were in Hz. Mass spectra were carried out using electrospray ionization techniques (ESI). Merck silica gel 60 PF₂₅₄ and 60 (0.063–0.2 mm) were used for preparative thin-layer chromatography and column chromatography, respectively. The reagent for TLC analysis was KMnO₄ in NaOH/K₂CO₃ aqueous solution, and the TLC was heated until the development of color.

The preparation of substrates 5a–5p is commented on in the Supplementary Materials. Compounds 5f [84], 5l [57], and 5m [57] have been previously reported. Compounds 5n and 5o are commercial products. The synthesis of the aldehydes 6a–p and the lactones 7a–u is described below.

General Procedure for the synthesis of aldehydes by oxidative radical scission of N-substituted hydroxypyrrolidines. To a solution of the N-substituted hydroxypyrrolidine (1.0 mmol) in dry dichloromethane (20 mL), iodine (127.0 mg, 0.50 mmol) and PhI(OAc)₂ (644.0 mg, 2.0 mmol) were added. The resulting mixture was stirred for 30–90 min at 26 °C under irradiation with visible light (cool white LED lamp). Then, the reaction mixture was poured into 10% aqueous Na₂S₂O₃ (10 mL) and extracted with CH₂Cl₂. The organic layer was dried over sodium sulfate, filtered, and concentrated under vacuum. The residue was purified by chromatography on silica gel (hexanes/ethyl acetate) to give the scission products 2a−2p.

Methyl (2S)-N-(acetoxymethyl)-N-(phenylsulfonyl)-4-oxo-L-homoalanine (6a). Obtained from N-phenylsulfonyl-L-hydroxyproline 5a (228.0 mg, 0.80 mmol) according to the general pyrrolidine scission procedure. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 60:40), yielding aldehyde 6a (218.6 mg, 0.64 mmol, 80%) as a colorless viscous oil. [α]_D: −29 (c 0.45, CHCl₃). IR (CHCl₃) ν_max 3023, 1745, 1448, 1437 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 9.63 (s, 1H), 7.88 (br d, J = 9.0 Hz, 2H), 7.62 (t, J = 7.8 Hz, 1H), 7.53 (t, J = 7.4 Hz, 2H), 5.60 (d, J = 12.4 Hz, 1H), 5.39 (d, J = 12.4 Hz, 1H), 5.08 (t, J = 6.9 Hz, 1H), 3.58 (s, 3H), 3.21 (ddd, J = 18.3, 7.4, 0.8 Hz, 1H), 2.88 (ddd, J = 18.1, 6.5, 1.0 Hz, 1H), 1.94 (s, 3H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 197.4 (CH), 170.1 (C), 169.7 (C), 139.7 (C), 133.5 (CH), 129.1 (2 × CH), 127.8 (2 × CH), 71.1 (CH₂), 54.4 (CH), 53.0 (CH₃), 44.6 (CH₂), 20.8 (CH₃). HRMS (ESI) calculated for C₁₅H₂₁NO₈SNa [M + MeOH + Na]⁺ 398.0886, found 398.0878. Anal. Calcd for C₁₄H₁₇NO₇S: C, 48.97; H, 4.99; N, 4.08; S, 9.34. Found: C, 49.09; H, 4.98; N, 4.04; S, 9.04.

Methyl (2S)-N-(acetoxymethyl)-N-(toluenesulfonyl)-4-oxo-L-homoalanine (6b). Obtained from N-toluensulfonyl-L-hydroxyproline 5b (228.0 mg, 0.80 mmol) according to the general pyrrolidine scission procedure. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 60:40), yielding aldehyde 6b (156.5 mg, 0.44 mmol, 73%) as a yellow oil. [α]_D: −39 (c 0.86, CHCl₃). IR (CHCl₃) ν_max 3029, 1744, 1365, 1352 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 9.62 (s, 1H), 7.75 (d, J = 8.4 Hz, 2H), 7.31 (d, J = 7.9 Hz, 2H), 5.60 (d, J = 12.4 Hz, 1H), 5.37 (d, J = 12.2 Hz, 1H), 5.05 (t, J = 6.9 Hz, 1H), 3.60 (s, 3H), 3.19 (dd, J = 18.1, 7.2 Hz, 1H), 2.86 (dd, J = 18.1, 6.4 Hz, 1H), 2.43 (s, 3H), 1.95 (s, 3H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 197.5 (CH), 170.2 (C), 169.8 (C), 144.5 (C), 136.8 (C), 129.7 (2 × CH), 127.9 (2 × CH), 71.2 (CH₂), 54.4 (CH), 53.0 (CH₃), 44.7 (CH₂), 21.7 (CH₃), 20.9 (CH₃). HRMS (ESI) calculated for C₁₆H₂₃NO₈SNa [M + MeOH + Na]⁺ 412.1042, found 412.1050. Anal. Calcd for C₁₅H₁₉NO₇S: C, 50.41; H, 5.36; N, 3.92; S, 8.97. Found: C, 50.28; H, 5.25; N, 4.24; S, 8.92.

Methyl (2S)-N-(acetoxymethyl)-N-(p-chlorophenylsulfonyl)-4-oxo-L-homoalanine (6c). Obtained from N-(p-chlorophenylsulfonyl)-L-hydroxyproline 5c (255.2 mg, 0.80 mmol) according to the general pyrrolidine scission procedure. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 70:30), yielding aldehyde 6c (220.4 mg, 0.58 mmol, 73%) as a colorless oil. [α]_D: −24 (c 0.34, CHCl₃). IR (CHCl₃) ν_max 3022, 1746, 1397, 1356, 1167 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 9.65 (br s, 1H), 7.83 (br d, J = 8.3 Hz, 2H), 7.50 (br d, J = 8.9 Hz, 2H), 5.58 (d, J = 11.8 Hz, 1H), 5.36 (d, J = 12.5 Hz, 1H), 5.07 (t, J = 6.9 Hz, 1H), 3.60 (s, 3H), 3.23 (ddd, J = 18.2, 7.1, 0.8 Hz, 1H), 2.91 (ddd, J = 18.2, 6.5, 1.0 Hz, 1H), 1.94 (s, 3H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 197.2 (CH), 170.1 (C), 169.7 (C), 140.1 (C), 138.2 (C), 129.38 (2 × CH), 129.36 (2 × CH), 70.9 (CH₂), 54.5 (CH), 53.1 (CH₃), 44.6 (CH₂), 20.8 (CH₃). HRMS (ESI) calculated for C₁₅H₂₀ClNO₈SNa [M + MeOH + Na]⁺ 432.0496, found 432.0508. Anal. Calcd for C₁₄H₁₆ClNO₇S: C, 44.51; H, 4.27; N, 3.71; S, 8.49. Found: C, 44.31; H, 3.88; N, 3.60; S, 8.30.

Methyl (2S)-N-(acetoxymethyl)-N-(p-iodophenylsulfonyl)-4-oxo-L-homoalanine (6d). Obtained from N-(p-iodophenylsulfonyl)-L-hydroxyproline 5d (328.8 mg, 0.80 mmol) according to the general pyrrolidine scission procedure. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 60:40), yielding aldehyde 6d (320.6 mg, 0.68 mmol, 86%) as a yellow oil. [α]_D: −39 (c 0.34, CHCl₃). IR (CHCl₃) ν_max 3027, 2955, 2847, 1747, 1570 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 9.66 (s, 1H), 7.89 (br b, J = 8.7 Hz, 2H), 7.60 (br b, J = 8.7 Hz, 2H), 5.58 (d, J = 12.4 Hz, 1H), 5.37 (d, J = 12.4 Hz, 1H), 5.06 (t, J = 6.9 Hz, 1H), 3.61 (s, 3H), 3.23 (dd, J = 17.9, 7.2 Hz, 1H), 2.92 (dd, J = 18.2, 6.5 Hz, 1H), 1.95 (s, 3H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 197.2 (CH), 170.1 (C), 169.6 (C), 139.4 (C), 138.4 (2 × CH), 129.2 (2 × CH), 101.1 (C), 70.9 (CH₂), 54.5 (CH), 53.1 (CH₃), 44.6 (CH₂), 20.8 (CH₃). HRMS (ESI) calculated for C₁₅H₂₀INO₈SNa [M + MeOH + Na]⁺ 523.9852, found 523.9852. Anal. Calcd for C₁₄H₁₆INO₇S: C, 35.83; H, 3.44; N, 2.98; S, 6.83. Found: C, 35.57; H, 3.47; N, 2.93; S, 6.78.

Methyl (2S)-N-(acetoxymethyl)-N-(p-nitrophenylsulfonyl)-4-oxo-L-homoalanine (6e). Obtained from N-(p-nitrophenylsulfonyl)-L-hydroxyproline 5e (204.6 mg, 0.60 mmol) according to the general pyrrolidine scission procedure. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 60:40), yielding aldehyde 6e (171.4 mg, 0.44 mmol, 71%) as a yellow oil. [α]_D: −17 (c 0.23, CHCl₃). IR (CHCl₃) ν_max 3019, 1749, 1535, 1350 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 9.60 (s, 1H), 8.30 (br b, J = 8.5 Hz, 2H), 8.04 (br b, J = 8.5 Hz, 2H), 5.57 (d, J = 12.4 Hz, 1H), 5.33 (d, J = 12.4 Hz, 1H), 5.06 (t, J = 6.9 Hz, 1H), 3.56 (s, 3H), 3.20 (dd, J = 18.4, 6.6 Hz, 1H), 2.94 (dd, J = 18.5, 7.1 Hz, 1H), 1.87 (s, 3H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 196.9 (CH), 170.0 (C), 169.5 (C), 150.5 (C), 145.4 (C), 129.4 (2 × CH), 124.2 (2 × CH), 70.5 (CH₂), 54.6 (CH), 53.2 (CH₃), 44.5 (CH₂), 20.7 (CH₃). HRMS (ESI) calculated for C₁₅H₂₀N₂O₁₀SNa [M + MeOH + Na]⁺ 443.0736, found 443.0730. Anal. Calcd for C₁₄H₁₆N₂O₉S: C, 43.30; H, 4.15; N, 7.21; S, 8.26. Found: C, 43.26; H, 4.22; N, 7.31; S, 8.47.

Methyl (2S)-N-(acetoxymethyl)-N-benzoyl-4-oxo-L-homoalanine (6f). Obtained from N-benzoyl-L-hydroxyproline 5f (249.1 mg, 1.0 mmol) according to the general pyrrolidine scission procedure. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 60:40), yielding aldehyde 6f (270.8 mg, 0.88 mmol, 88%) as a yellow oil, whose characterization data were already reported [56]. [α]_D: −76 (c 0.42, CHCl₃); ¹H NMR (500 MHz, CDCl₃, 26 °C): δ_H 2.12 (3H, s), 3.25 (1H, dd, J = 7.6, 18.8 Hz), 3.51 (1H, br d, J = 16.7 Hz), 3.78 (3H, s), 4.95 (1H, dd, J = 5.3, 7.7 Hz), 5.43 (2H, br s), 7.35–7.53 (5H, m), 9.80 (1H, s); HRMS (ESI-TOF): calcd for C₁₅H₁₇NO₆Na (M⁺ + Na), 330.0954; found, 330.0952.

Methyl (2S)-N-(acetoxymethyl)-N-(p-fluorobenzoyl)-4-oxo-L-homoalanine (6g). Obtained from N-(p-fluorobenzoyl)-L-hydroxyproline 5g (267.1 mg, 1.0 mmol) according to the general pyrrolidine scission procedure. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 60:40), yielding aldehyde 6g (277.7 mg, 0.85 mmol, 85%) as a yellow oil. [α]_D: −56 (c 0.39, CHCl₃). IR (CHCl₃) ν_max 3021, 1745, 1658, 1604 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 55 °C) δ_H 9.81 (s, 1H), 7.55–7.50 (m, 2H), 7.10 (t, J_H,H = 8.6, J_H,F = 8.6, Hz, 2H), 5.43 (d, J = 11.6 Hz, 1H), 5.38 (d, J = 11.6 Hz, 1H), 4.95 (dd, J = 7.9, 5.2 Hz, 1H), 3.77 (s, 3H), 3.47 (dd, J = 18.6, 5.2 Hz, 1H), 3.20 (dd, J = 18.6, 8.0 Hz, 1H), 2.10 (s, 3H). ¹³C RMN (125.7 MHz, CDCl₃, 55 °C) δ_C 198.4 (CH), 171.7 (C), 170.4 (C), 170.0 (C), 164.4 (C, d, J_CF = 252.7 Hz), 130.6 (C, d, J_CF = 3.68 Hz), 130.1 (2 × CH, d, J_CF = 8.72 Hz), 115.8 (2 × CH, d, J_CF = 22.1 Hz), 74.1 (CH₂), 55.2 (CH), 52.9 (CH₃), 44.2 (CH₂), 20.9 (CH₃). HRMS (ESI) calculated for C₁₆H₂₀FNO₇Na [M + MeOH + Na]⁺ 380.1121, found 380.1126. Anal. Calcd for C₁₅H₁₆FNO₆: C, 55.39; H, 4.96; N, 4.31. Found: C, 55.44; H, 5.31; N, 4.24.

Methyl (2S)-N-(acetoxymethyl)-N-(p-chlorobenzoyl)-4-oxo-L-homoalanine (6h). Obtained from N-(p-chlorobenzoyl)-L-hydroxyproline 5h (226.5 mg, 0.80 mmol) according to the general pyrrolidine scission procedure. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 60:40), yielding aldehyde 6h (222.7 mg, 0.65 mmol, 82%) as a yellow oil. [α]_D: −63 (c 0.36, CHCl₃). IR (CHCl₃) ν_max 3021, 1746, 1656, 1598 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 9.80 (s, 1H), 7.45 (br d, J = 8.6, 2H), 7.40 (br d, J = 8.6 Hz, 2H), 5.44–5.35 (br b, 2H), 4.91 (dd, J = 8.2, 5.0 Hz, 1H), 3.76 (s, 3H), 3.54–3.45 (m, 1H), 3.24 (dd, J = 19.1, 8.1 Hz, 1H), 2.11 (s, 3H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 198.8 (CH), 171.7 (C), 170.6 (C), 170.0 (C), 137.3 (C), 132.7 (C), 129.1 (2 × CH), 128.9 (2 × CH), 74.2 (CH₂), 55.0 (CH), 53.0 (CH₃), 44.1 (CH₂), 20.9 (CH₃). HRMS (ESI) calculated for C₁₆H₂₀NO₇ClNa [M + MeOH + Na]⁺ 396.0826, found 396.0821. Anal. Calcd for C₁₅H₁₆NO₆Cl: C, 52.72; H, 4.72; N, 4.10. Found: C, 52.15; H, 4.51; N, 4.00.

Methyl (2S)-N-(acetoxymethyl)-N-(p-iodobenzoyl)-4-oxo-L-homoalanine (6i). Obtained from N-(p-iodobenzoyl)-L-hydroxyproline 5i (300.0 mg, 0.80 mmol) according to the general pyrrolidine scission procedure. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 70:30), yielding aldehyde 6i (251.0 mg, 0.58 mmol, 73%) as a colorless oil. [α]_D: −66 (c 0.36, CHCl₃). IR (CHCl₃) ν_max 3021, 1745, 1658, 1587 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 9.81 (s, 1H), 7.78 (br b, J = 8.4 Hz, 2H), 7.23 (d, J = 8.4 Hz, 2H), 5.38 (br b, 2H), 4.91 (dd, J = 8.1, 4.9 Hz, 1H), 3.76 (s, 3H), 3.54–3.46 (m, 1H), 3.25 (dd, J = 18.9, 8.1 Hz, 1H), 2.11 (s, 3H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 198.8 (CH), 171.9 (C), 170.6 (C), 169.9 (C), 137.8 (2 × CH), 133.7 (C), 129.2 (2 × CH), 97.7 (C), 74.2 (CH₂), 55.0 (CH), 53.0 (CH₃), 44.1 (CH₂), 20.9 (CH₃). HRMS (ESI) calculated for C₁₆H₂₀INO₇Na [M + MeOH + Na]⁺ 488.0182, found 488.0181. Anal. Calcd for C₁₅H₁₆INO₆: C, 41.59; H, 3.72; N, 3.23. Found: C, 41.77; H, 3.86; N, 3.53.

Methyl (2S)-N-(acetoxymethyl)-N-(p-nitrobenzoyl)-4-oxo-L-homoalanine (6j). Obtained from N-(p-nitrobenzoyl)-L-hydroxyproline 5j (235.3 mg, 0.80 mmol) according to the general pyrrolidine scission procedure. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 60:40), yielding aldehyde 6j (245.4 mg, 0.70 mmol, 87%) as a colorless oil. [α]_D: −63 (c 0.33, CHCl₃). IR (CHCl₃) ν_max 3022, 1748, 1662, 1528, 1347 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 9.83 (s, 1H), 8.29 (d, J = 8.6 Hz, 2H), 7.66 (d, J = 8.7 Hz, 2H), 5.39 (d, J = 11.8 Hz, 1H), 5.32 (d, J = 11.8 Hz, 1H), 4.94 (dd, J = 8.5, 4.6 Hz, 1H), 3.78 (s, 3H), 3.53 (dd, J = 19.1, 4.6 Hz, 1H), 3.33 (dd, J = 19.2, 8.5 Hz, 1H), 2.11 (s, 3H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 198.8 (CH), 170.7 (C), 170.6 (C), 169.6 (C), 149.1 (C), 140.4 (C), 128.7 (2 × CH), 123.9 (2 × CH), 73.7 (CH₂), 55.0 (CH), 53.2 (CH₃), 44.0 (CH₂), 20.9 (CH₃). HRMS (ESI) calculated for C₁₆H₂₀N₂O₉Na [M + MeOH + Na]⁺ 407.1066, found 407.1066. Anal. Calcd for C₁₅H₁₆N₂O₈: C, 51.14; H, 4.58; N, 7.95. Found: C, 51.44; H, 4.73; N, 7.60.

Methyl (2S)-N-(acetoxymethyl)-N-(3,5-dinitrobenzoyl)-4-oxo-L-homoalanine (6k). Obtained from N-(3,5-dinitrobenzoyl)-L-hydroxyproline 5k (339.1 mg, 1.00 mmol) according to the general pyrrolidine scission procedure. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 60:40), yielding aldehyde 6k (309.4 mg, 0.78 mmol, 78%) as a yellow oil. [α]_D: −54 (c 0.35, CHCl₃). IR (CHCl₃) ν_max 1749, 1670, 1548, 1344 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 9.84 (s, 1H), 9.15 (t, J = 2.1 Hz, 1H), 8.72 (br b, 2H), 5.42 (d, J = 11.9 Hz, 1H), 5.29 (d, J = 11.9 Hz, 1H), 5.04–4.96 (m, 1H), 3.81 (s, 3H), 3.59–3.49 (m, 1H), 3.38 (dd, J = 19.2, 9.0 Hz, 1H), 2.16 (s, 3H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 198.4 (CH), 170.5 (C), 169.3 (C), 168.2 (C), 148.5 (2 × C), 137.8 (C), 128.1 (2 × CH), 120.7 (CH), 73.5 (CH₂), 55.4 (CH), 53.4 (CH₃), 43.9 (CH₂), 20.7 (CH₃). HRMS (ESI) calculated for C₁₆H₁₉N₃O₁₁Na [M + MeOH + Na]⁺ 452.0917, found 452.0922. Anal. Calcd for C₁₅H₁₅N₃O₁₀: C, 45.35; H, 3.81; N, 10.58. Found: C, 45.59; H, 3.56; N, 10.86.

Methyl (2S)-N-(acetyl)-N-(acetoxymethyl)-4-oxo-L-homoalanine (6l). Obtained from N-(acetyl)-L-hydroxyproline 5l (112.1 mg, 0.60 mmol) according to the general pyrrolidine scission procedure. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 50:50), yielding aldehyde 6l (105.7 mg, 0.43 mmol, 72%) as a yellow oil. [α]_D: −80 (c 0.33, CHCl₃). IR (CHCl₃) ν_max 3023, 3012, 1744, 1673 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 9.74 (s, 1H), 5.55 (d, J = 12.0 Hz, 1H), 5.38 (d, J = 12.0 Hz, 1H), 4.79 (dd, J = 7.9, 4.9 Hz, 1H), 3.69 (s, 3H), 3.41 (dd, J = 18.9, 4.9 Hz, 1H), 3.12 (dd, J = 18.9, 8.2 Hz, 1H), 2.22 (s, 3H), 2.09 (s, 3H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 199.2 (CH), 172.0 (C), 170.7 (C), 170.2 (C), 73.5 (CH₂), 55.2 (CH₂), 52.9 (CH₃), 44.5 (CH₂), 21.6 (CH₃), 20.9 (CH₃). HRMS (ESI) calculated for C₁₁H₁₉NO₇Na [M + MeOH + Na]⁺ 300.1059, found 300.1054. Anal. Calcd for C₁₀H₁₅NO₆: C, 48.98; H, 6.17; N, 5.71. Found: C, 48.71; H, 6.18; N, 5.62.

Methyl (2S)-N-(acetoxymethyl)-N-[N-(benzyloxycarbonyl)phenylalanyl]-4-oxo -L-homoalanine (6m). Obtained from N-[N-(benzyloxycarbonyl)phenylalanyl]- L-hydroxyproline 5m (298.3 mg, 0.70 mmol) according to the general pyrrolidine scission procedure. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 50:50), yielding aldehyde 6m (238.6 mg, 0.49 mmol, 70%) as a yellow oil. [α]_D: −26 (c 0.36, CHCl₃). IR (CHCl₃) ν_max 3301, 1716, 1639, 1524, 1435 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 9.64 (br b, 1H), 7.38–7.04 (m, 10H), 5.47 (d, J = 8.4 Hz, 1H), 5.37 (d, J = 12.2 Hz, 1H), 5.19 (d, J = 12.2 Hz, 1H), 5.12 (d, J = 12.3 Hz, 1H), 5.08 (d, J = 12.4 Hz, 1H), 5.08–5.03 (m, 1H), 4.70 (dd, J = 7.6, 5.1 Hz, 1H), 3.63 (s, 3H), 3.34 (dd, J = 18.9, 5.2 Hz, 1H), 3.04–2.96 (m, 2H), 2.81 (dd, J = 18.8, 7.6 Hz, 1H), 2.01 (s, 3H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 198.5 (CH), 173.2 (C), 170.6 (C), 169.7 (C), 155.5 (C), 136.4 (C), 135.7 (C), 129.6 (2 × CH), 128.7 (2 × CH), 128.7 (2 × CH), 128.3 (CH), 128.2 (2 × CH), 127.4 (CH), 72.1 (CH₂), 67.1 (CH₂), 55.7 (CH), 52.8 (CH₃), 52.4 (CH), 44.0 (CH₂), 40.0 (CH₂), 20.7 (CH₃). HRMS (ESI) calculated for C₂₆H₃₂N₂O₉Na [M + MeOH + Na]⁺ 507.1743, found 507.1740. Anal. Calcd for C₂₅H₂₈N₂O₈: C, 61.98; H, 5.83; N, 5.78. Found: C, 61.81; H, 6.02; N, 5.41.

Methyl (2S)-N-(acetoxymethyl)-N-(terc-butoxycarbonyl)-4-oxo-L-homoalanine (6n). Obtained from N-(tert-butoxycarbonyl)-L-hydroxyproline 5n (147.2 mg, 0.60 mmol) according to the general pyrrolidine scission procedure. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 80:20), yielding aldehyde 6n (126.0 mg, 0.42 mmol, 70%) as a yellow oil whose characterization data were already reported [56]. [α]_D: −80 (c 0.33, CHCl₃); ¹H NMR (500 MHz, CDCl₃, 70 °C) rotamer mixture at 26 °C, one visible rotamer at 70 °C: δ_H 1.47 (9H, s), 2.05 (3H, s/s), 2.95 (1H, m), 3.31 (1H, dd, J = 6.1, 18 Hz), 3.73 (3H, s), 4.83 (1H, m), 5.41 (2H, br s), 9.76 (1H, s); HRMS (ESI-TOF): calcd for C₁₄H₂₅NO₈Na (M⁺ + Na + MeOH), 358.1478; found, 358.1467.

Methyl (2S)-N-(acetoxymethyl)-N-(benzyloxycarbonyl)-4-oxo-L-homoalanine (6o). Obtained from N-(benzyloxycarbonyl)-L-hydroxyproline 5o (223.3 mg, 0.80 mmol) according to the general pyrrolidine scission procedure. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 70:30), yielding aldehyde 6o (152.2 mg, 0.45 mmol, 56%) as a colorless oil. [α]_D: −70 (c 0.36, CHCl₃). IR (CHCl₃) ν_max 1726, 1437, 1421, 1367 cm⁻¹. ¹H NMR (500 MHz, CD₃CN, 70 °C) δ_H 9.68 (s, 1H), 7.42–7.32 (m, 5H), 5.43 (d, J = 11.3 Hz, 1H), 5.41 (d, J = 11.2 Hz, 1H), 5.16 (br b, 2H), 4.92 (t, J = 6.7 Hz, 1H), 3.61 (s, 3H), 3.24 (dd, J = 18.1, 6.4 Hz, 1H), 2.95 (dd, J = 18.2, 7.0 Hz, 1H), 1.99 (s, 3H). ¹³C RMN (125.7 MHz, CD₃CN, 70 °C) δ_C 200.4 (CH), 171.8 (C), 171.7 (C), 156.4 (C), 137.6 (C), 129.8 (2 × CH), 129.5 (CH), 129.1 (2 × CH), 73.4 (CH₂), 69.1 (CH₂), 56.6 (CH), 53.4 (CH₃), 45.5 (CH₂), 21.2 (CH₃). HRMS (ESI) calculated for C₁₇H₂₃NO₈Na [M + MeOH + Na]⁺ 392.1321, found 392.1319. Anal. Calcd for C₁₆H₁₉NO₇: C, 56.97; H, 5.68; N, 4.15. Found: C, 56.87; H, 5.87; N, 4.08.

Methyl (2S)-N-(acetoxymethyl)-N-(phenyloxycarbonyl)-4-oxo-L-homoalanine (6p). Obtained from N-(phenyloxycarbonyl)-L-hydroxyproline 5p (265.1 mg, 1.00 mmol) according to the general pyrrolidine scission procedure. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 70:30), yielding aldehyde 6p (256.4 mg, 0.79 mmol, 79%) as a yellow oil. [α]_D: −89 (c 0.36, CHCl₃). IR (CHCl₃) ν_max 1731, 1599, 1417, 1288, 1198 cm⁻¹. ¹H NMR (500 MHz, CD₃CN, 70 °C) δ_H 9.76 (s, 1H), 7.42 (br t, J = 7.8 Hz, 2H), 7.28 (t, J = 7.7 Hz, 1H), 7.14 (d, J = 7.9 Hz, 2H), 5.64–5.47 (m, 2H), 5.09–4.96 (m, 1H), 3.74 (s, 3H), 3.34 (dd, J = 18.2, 5.7 Hz, 1H), 3.09 (dd, J = 17.9, 5.6 Hz, 1H), 2.02 (s, 3H). ¹³C RMN (125.7 MHz, CD₃CN, 70 °C) δ_C 200.3 (CH), 171.8 (C), 171.6 (C), 155.1 (C), 152.4 (C), 130.7 (2 × CH), 127.1 (CH), 122.7 (2 × CH), 73.5 (CH₂), 56.9 (CH), 53.6 (CH₃), 45.4 (CH₂), 21.2 (CH₃). HRMS (ESI) calculated for C₁₆H₂₁NO₈Na [M + MeOH + Na]⁺ 378.1165, found 378.1168. Anal. Calcd for C₁₅H₁₇NO₇: C, 55.73; H, 5.30; N, 4.33. Found: C, 55.61; H, 5.43; N, 4.43.

General procedure for the preparation of homoserine lactones.

Method A: A solution of the 4-oxo-L-homoalanine derivative (0.20 mmol) in dry methanol (3.0 mL), at room temperature, was treated with NaBH₄ (9.8 mg, 0.26 mmol, 1.3 equiv.). The reaction mixture was stirred at 45 °C for 4 h. Then, the solvent was removed under vacuum, and the residue was poured into water and extracted with EtOAc. The organic phase was dried over anhydrous sodium sulfate, filtered, and concentrated under vacuum. The crude oil obtained was dissolved in dichloromethane (3.0 mL), and Et₃N (100 μL) was added. The mixture was stirred at 40 °C for 1 h, and then the solvent was removed under vacuum. The residue was purified by radial chromatography on silica gel (n-hexane/EtOAc) to obtain the corresponding α-amino lactones.

Method B: To a solution of the 4-oxo-L-homoalanine derivative (0.20 mmol) in dry dichloromethane (4.0 mL), boron trifluoride diethyleterate (50 μL, 57.5 mg, 0.40 mmol, 2.0 equiv.) and triethylsilane (80 μL, 58.0 mg, 0.57 mmol, 2.5 equiv.) were added. The reaction mixture was stirred at 26 °C for 16 h under a nitrogen atmosphere. Then, Et₃N (100 μL) was added, and stirring was continued for 2 h. The mixture was concentrated under vacuum and the residue was purified by radial chromatography on silica gel (n-hexane/EtOAc mixtures), yielding the corresponding N-alkyl-α-amino lactones.

(2S)-N-(Benzenesulfonyl)homoserine lactone (7a). Obtained from aldehyde 6a (68.6 mg, 0.20 mmol) according to method A of the general procedure for the preparation of lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 70:30), yielding lactone 7a (30.4 mg, 0.13 mmol, 63%) as a colorless oil, which was known [85] but not completely characterized: [α]_D: −2 (c 0.25, CHCl₃). IR (CHCl₃) ν_max 3330, 1785, 1602, 1349, 1169 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 7.91 (d, J = 7.8 Hz, 2H), 7.62 (t, J = 7.5 Hz, 1H), 7.54 (t, J = 7.8 Hz, 2H), 4.41 (t, J = 9.1 Hz, 1H), 4.19 (ddd, J = 11.7, 9.4, 5.7 Hz, 1H), 3.97 (dd, J = 11.6, 8.4 Hz, 1H), 2.69 (ddd, J = 12.7, 8.5, 5.6 Hz, 1H), 2.31–2.21 (m, 1H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 174.2 (C), 139.1 (C), 133.4 (CH), 129.5 (2 × CH), 127.4 (2 × CH), 66.2 (CH₂), 51.9 (CH), 31.3 (CH₂). HRMS (ESI) calculated for C₁₀H₁₁NO₄SNa [M + Na]⁺ 264.0306, found 264.0304. Anal. Calcd for C₁₀H₁₁NO₄S: C, 49.78; H, 4.60; N, 5.81; S, 13.29. Found: C, 49.62; H, 4.70; N, 5.80; S, 13.54.

(2S)-N-(p-Toluenesulfonyl)homoserine lactone (7b). Obtained from aldehyde 6b (38.2 mg, 0.11 mmol) according to method A for the preparation of homoserine lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 70:30), yielding lactone 7b (17.7 mg, 0.07 mmol, 65%) as a crystalline solid whose characterization data were already reported [49], but since the deuterated solvent is different (CDCl3/d-DMSO) our data are given herein. [α]_D: −2 (c 0.48, CHCl₃). ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 7.81–7.77 (m, 2H), 7.33 (d, J = 7.8 Hz, 2H), 5.30 (d, J = 2.1 Hz, 1H), 4.41 (t, J = 9.2 Hz, 1H), 4.18 (ddd, J = 11.7, 9.4, 5.6 Hz, 1H), 3.92 (ddd, J = 11.9, 8.3, 3.8 Hz, 1H), 2.69 (m, 1H), 2.43 (s, 3H), 2.27 (m, 1H). HRMS (ESI) calculated for C₁₁H₁₃NO₄SNa [M + Na]⁺ 278.0463, found 278.0461.

(2S)-N-(p-Chlorophenylsulfonyl)homoserine lactone (7c). Obtained from aldehyde 6c (75.4 mg, 0.20 mmol) according to method A for the preparation of homoserine lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 60:40), yielding lactone 7c (16.0 mg, 0.06 mmol, 29%) as a colorless oil. [α]_D: 3 (c 0.33, (CH₃)₂CO). IR (ATR) ν_max 3022, 1746, 1356, 1232, 1167 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 7.85 (br d, J = 8.5 Hz, 2H), 7.51 (br d, J = 8.6 Hz, 2H), 5.60–5.40 (br b, 1H), 4.43 (t, J = 9.1 Hz, 1H), 4.20 (ddd, J = 11.6, 9.5, 5.6 Hz, 1H), 4.00 (dd, J = 11.7, 8.4 Hz, 1H), 2.70 (dddd, J = 12.8, 8.3, 5.6, 1.2 Hz, 1H), 2.31–2.21 (m, 1H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 174.1 (C), 140.0 (C), 137.8 (C), 129.8 (2 × CH), 128.9 (2 × CH), 66.2 (CH₂), 52.0 (CH), 31.3 (CH₂, 4-C). HRMS (ESI) calculated for C₁₀H₁₀ClNO₄SNa [M + Na]⁺ 297.9917, found 297.9919.

(2S)-N-(p-Iodophenylsulfonyl)homoserine lactone (7d). Obtained from aldehyde 6d (93.8 mg, 0.20 mmol) according to method A for the preparation of homoserine lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 70:30), yielding lactone 7d (38.2 mg, 0.10 mmol, 52%) as a crystalline solid: mp 141–143 °C (from n-hexane/EtOAc); [α]_D: +4 (c 0.40, (CH₃)₂CO). IR (ATR) ν_max 3238, 2921, 1780, 1337, 1162 cm⁻¹. ¹H NMR (500 MHz, (CD₃)₂CO, 26 °C) δ_H 8.00 (br d, J = 8.6 Hz), 7.71 (d, J = 8.7 Hz, 2H), 4.41 (dd, J = 11.6, 8.6 Hz, 1H), 4.34 (td, J = 8.9, 1.4 Hz, 1H), 4.24 (ddd, J = 11.1, 9.1, 5.9 Hz, 1H), 2.53 (dddd, J = 12.5, 8.6, 5.9, 1.4 Hz, 1H), 2.20–2.10 (m, 1H). ¹³C RMN (125.7 MHz, (CD₃)₂CO, 26 °C) δ_C 174.6 (C), 142.3 (C), 139.2 (2 × CH), 129.5 (2 × CH), 100.1 (C), 66.1 (CH₂), 52.7 (CH), 31.2 (CH₂). HRMS (ESI) calculated for C₁₀H₁₀INO₄SNa [M + Na]⁺ 389.9273, found 389.9269.

(2S)-N-(p-Nitrophenylsulfonyl)homoserine lactone (7e).Obtained from aldehyde 6e (77.6 mg, 0.20 mmol) according to method A for the preparation of homoserine lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 70:30), yielding lactone 7e (32.0 mg, 0.11 mmol, 56%) as a crystalline solid: mp 162–164 °C (from n-hexane/EtOAc); [α]_D: +1 (c 0.33, (CH₃)₂CO). IR (ATR) ν_max 3297, 1781, 1534, 1347, 1169 cm⁻¹. ¹H NMR (500 MHz, (CD₃)₂CO, 26 °C) δ_H 8.43 (br d, J = 9.0 Hz, 2H), 8.20 (br d, J = 9.1 Hz, 2H), 4.54 (dd, J = 11.6, 8.5 Hz, 1H), 4.36 (ddd, J = 8.9, 8.9, 1.4 Hz, 1H), 4.26 (ddd, J = 11.1, 9.1, 5.9 Hz, 1H), 2.59 (dddd, J = 12.5, 8.6, 6.0, 1.4 Hz, 1H), 2.25–2.17 (m, 1H). ¹³C RMN (125.7 MHz, (CD₃)₂CO, 26 °C) δ_C 174.6 (C), 151.1 (C), 148.2 (C), 129.4 (2 × CH), 125.2 (2 × CH), 66.1 (CH₂), 52.8 (CH), 31.1 (CH₂). HRMS (ESI) calculated for C₁₀H₁₀N₂O₆SNa [M + Na]⁺ 309.0157, found 309.0158. Anal. Calcd for C₁₀H₁₀N₂O₆S: C, 41.96; H, 3.52; N, 9.79; S, 11.20. Found: C, 41.69; H, 3.53; N, 9.51; S, 10.94.

(2S)-N-(Benzoyl)homoserine lactone (7f). Obtained from aldehyde 6f (30.7 mg, 0.10 mmol) according to method A for the preparation of homoserine lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 50:50), yielding lactone 7f (13.7 mg, 0.01 mmol, 67%) as a crystalline solid: mp 126–128 °C (from n-hexane/EtOAc); [α]_D: +15 (c 0.34, CHCl₃). IR (CHCl₃) ν_max 3430, 3017, 1779, 1667, 1514, 1486 cm⁻¹. ¹H NMR (500 MHz, CD₃CN, 70 °C) δ_H 7.83–7.79 (m, 2H), 7.56 (br t, J = 7.5 Hz, 1H), 7.48 (br t, J = 7.5 Hz, 2H), 4.71 (ddd, J = 11.0, 9.2, 7.9 Hz, 1H), 4.45 (ddd, J = 9.0, 9.0, 2.0 Hz, 1H), 4.28 (ddd, J = 10.4, 9.0, 6.6 Hz, 1H), 2.60–2.52 (m, 1H), 2.46–2.36 (m, 1H). ¹³C RMN (125.7 MHz, CD₃CN, 70 °C) δ_C 176.3 (C), 168.3 (C), 135.3 (C), 133.0 (CH), 129.9 (2 × CH), 128.4 (2 × CH), 67.0 (CH₂), 50.3 (CH), 29.7 (CH₂). HRMS (ESI) calculated for C₁₁H₁₁NO₃Na [M + Na]⁺ 228.0637, found 228.0635. Anal. Calcd for C₁₁H₁₁NO₃: C, 64.38; H, 5.40; N, 6.83. Found: C, 64.36; H, 5.78; N, 6.58.

(2S)-N-(p-Fluorobenzoyl)homoserine lactone (7g). Obtained from aldehyde 6g (65.0 mg, 0.20 mmol) according to method A for the preparation of homoserine lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (hexanes/EtOAc, 50:50), yielding lactone 7g (30.0 mg, 0.13 mmol, 67%) as a crystalline solid: mp 152–154 °C (from n-hexane/EtOAc); [α]_D: +1 (c 0.81, (CH₃)₂CO). IR (CHCl₃) ν_max 3426, 1780, 1668, 1604, 1493 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 7.85–7.77 (m, 2H), 7.08 (br t, J_H,H = 8.5, J_H,F = 8.5 Hz, 2H), 7.01–6.92 (m, 1H), 4.81–4.72 (m, 1H), 4.52 (t, J = 9.1 Hz, 1H), 4.35 (ddd, J = 11.1, 9.2, 6.0 Hz, 1H), 2.94–2.85 (m, 1H), 2.35–2.24 (m, 1H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 176.1 (C), 166.8 (C), 165.5 (C, d, J_CF = 252.7 Hz), 129.6 (2 × CH, d, J_CF = 9.2 Hz), 129.2 (C, br s), 116.0 (2 × CH, d, J_CF = 22.5 Hz), 66.5 (CH₂), 49.8 (CH), 30.4 (CH₂). HRMS (ESI) calculated for C₁₁H₁₀FNO₃Na [M + Na]⁺ 246.0542, found 246.0541. Anal. Calcd for C₁₁H₁₀FNO₃: C, 59.19; H, 4.52; N, 6.28. Found: C, 59.23; H, 4.87; N, 6.00.

(2S)-N-(p-Chlorobenzoyl)homoserine lactone (7h). Obtained from aldehyde 6h (68.2 mg, 0.20 mmol) according to method A for the preparation of homoserine lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (CH₂Cl₂/MeOH, 98:2), yielding lactone 7h (18.8 mg; 0.08 mmol; 40%) as an amorphous solid. [α]_D: −1 (c 0.30, (CH₃)₂CO). IR (ATR) ν_max 3409, 1764, 1662, 1528, 1485 cm⁻¹. ¹H NMR (500 MHz, (CD₃)₂CO, 26 °C) δ_H 8.28 (br d, J = 4.6 Hz, 1H), 7.93 (br d, J = 8.9 Hz, 2H), 7.52 (br d, J = 8.9 Hz, 2H), 4.88 (ddd, J = 11.0, 9.1, 8.0 Hz, 1H), 4.47 (ddd, J = 9.0, 9.0, 1.9 Hz, 1H), 4.36 (ddd, J = 10.4, 8.8, 6.4 Hz, 1H), 2.70–2.61 (m, 1H), 2.50–2.41 (m, 1H). ¹³C RMN (125.7 MHz, (CD₃)₂CO, 26 °C) δ_C 175.4 (C), 166.3 (C), 138.0 (C), 133.7 (C), 129.9 (2 × CH), 129.5 (2 × CH), 66.2 (CH₂), 49.7 (CH), 29.5 (CH₂). HRMS (ESI) calculated for C₁₁H₁₀ClNO₃Na [M + Na]⁺ 262.0247, found 262.0239. Anal. Calcd for C₁₁H₁₀ClNO₃: C, 55.13; H, 4.21; N, 5.84. Found: C, 54.97; H, 4.22; N, 5.67.

(2S)-N-(p-Iodobenzoyl)homoserine lactone (7i). Obtained from aldehyde 6i (86.6 mg, 0.20 mmol) according to method A for the preparation of homoserine lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (CH₂Cl₂/MeOH, 99:1), yielding lactone 7i (38.7 mg, 0.12 mmol, 59%) as a crystalline solid: mp 206–208 °C (from CH₂Cl₂/MeOH); [α]_D: −1 (c 0.33, (CH₃)₂CO). IR (CHCl₃) ν_max 3258, 1773, 1644, 15,485, 1540 cm⁻¹. ¹H NMR (500 MHz, (CD₃)₂CO, 26 °C) δ_H 8.27 (br d, J = 6.7 Hz, 1H), 7.89 (br d, J = 9.1 Hz, 2H), 7.70 (br d, J = 9.1 Hz, 2H), 4.88 (ddd, J = 11.0, 9.1, 8.0 Hz, 1H), 4.46 (ddd, J = 8.9, 8.9, 1.8 Hz, 1H), 4.36 (ddd, J = 10.5, 8.9, 6.5 Hz, 1H), 2.69–2.61 (m, 1H), 2.45 (m, 1H). ¹³C RMN (125.7 MHz, (CD₃)₂CO, 26 °C) δ_C 175.4 (C), 166.7 (C), 138.6 (2 × CH), 134.5 (C), 130.0 (2 × CH), 98.9 (C), 66.2 (CH₂), 49.7 (CH), 29.5 (CH₂). HRMS (ESI) calculated for C₁₁H₁₀INO₃Na [M + Na]⁺ 353.9603, found 353.9605.

(2S)-N-(p-Nitrobenzoyl)homoserine lactone (7j). Obtained from aldehyde 6j (70.4 mg, 0.20 mmol) according to method A for the preparation of homoserine lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (n-hexane/EtOAc, 40:60), yielding lactone 7j (26.5 mg, 0.11 mmol, 53%) as a crystalline solid: mp 200–202 °C (from n-hexane/EtOAc); [α]_D: −1 (c 0.34, (CH₃)₂CO). IR (ATR) ν_max 3309, 1742, 1650, 1598, 1519 cm⁻¹. ¹H NMR (500 MHz, (CD₃)₂CO, 26 °C) δ_H 8.53 (br b, J = 6.6 Hz, 1H), 8.35 (br d, J = 8.8 Hz, 2H), 8.16 (br d, J = 8.9 Hz, 2H), 4.93 (ddd, J = 11.1, 9.2, 7.9 Hz, 1H), 4.49 (ddd, J = 9.0, 8.9, 1.9 Hz, 1H), 4.38 (ddd, J = 10.6, 9.0, 6.5 Hz, 1H), 2.73–2.65 (m, 1H), 2.54–2.44 (m, 1H). ¹³C RMN (125.7 MHz, (CD₃)₂CO, 26 °C) δ_C 175.2 (C), 165.7 (C), 150.7 (C), 140.4 (C), 129.6 (2 × CH), 124.5 (2 × CH), 66.2 (CH₂), 49.9 (CH), 29.4 (CH₂). HRMS (ESI) calculated for C₁₁H₁₀N₂O₅Na [M + Na]⁺ 273.0487, found 273.0487. Anal. Calcd for C₁₁H₁₀N₂O₅: C, 52.80; H, 4.03; N, 11.20. Found: C, 53.07; H, 4.07; N, 10.83.

(2S)-N-(3,5-Dinitrobenzoyl)homoserine lactone (7k). Obtained from aldehyde 6k (79.4 mg, 0.20 mmol) according to method A for the preparation of homoserine lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (n-hexane/EtOAc, 50:50), yielding lactone 7k (20.2 mg, 0.07 mmol, 34%) as a crystalline solid: mp 216–218 °C (from n-hexane/EtOAc); [α]_D: −16 (c 0.23, (CH₃)₂CO). IR (ATR) ν_max 3335, 1763, 1666, 1541, 1344 cm⁻¹. ¹H NMR (500 MHz, (CD₃)₂CO, 26 °C) δ_H 9.11 (s, 3H), 8.93 (br b, 1H), 5.05–4.97 (m, 1H), 4.51 (ddd, J = 9.1, 9.0, 1.9 Hz, 1H), 4.41 (ddd, J = 10.6, 9.0, 6.4 Hz, 1H), 2.77–2.70 (m, 1H), 2.57–2.47 (m, 1H). ¹³C RMN (125.7 MHz, (CD₃)₂CO, 26 °C) δ_C 175.0 (C), 163.5 (C), 149.7 (2 × C), 137.8 (C), 128.3 (2 × CH), 122.0 (CH), 66.3 (CH₂), 50.2 (CH), 29.4 (CH₂). HRMS (ESI) calculated for C₁₁H₉N₃O₇Na [M + Na]⁺ 318.0338, found 318.0339.

(2S)-N-(Acetyl)homoserine lactone (7l). Obtained from aldehyde 6l (24.5 mg, 0.10 mmol) according to method A for the preparation of homoserine lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (n-hexane/EtOAc, 10:90), yielding lactone 7l (6.2 mg, 0.04 mmol, 43%) as a crystalline solid whose characterization data have been reported but using different conditions/d-solvent [86]. [α]_D: −12 (c 0.23, CH₃COCH₃). ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 6.45–6.30 (s.a, 1H), 4.58 (ddd, J = 11.7, 8.6, 6.1 Hz, 1H), 4.46 (ddd, J = 9.1, 9.1, 1.0 Hz, 1H), 4.27 (ddd, J = 11.3, 9.3, 6.0 Hz, 1H), 2.76–2.84 (m, 1H), 2.15 (m, 1H), 2.05 (s, 3H). HRMS (ESI) calculated for C₆H₉NO₃Na [M + Na]⁺ 166.0480, found 166.0476.

(2S)-N-[N-(Benzyloxycarbonyl)phenylalanyl]homoserine lactone (7m). Obtained from aldehyde 6m (96.8 mg, 0.20 mmol) according to method A for the preparation of homoserine lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (n-hexane/EtOAc, 40:60), yielding lactone 7m (43.7 mg, 0.12 mmol, 57%) as a crystalline solid: mp 124–126 °C (from n-hexane/EtOAc); [α]_D: −3 (c 0.88, CHCl₃). IR (CHCl₃) ν_max 3447, 3011, 1785, 1672, 1509 cm⁻¹. ¹H NMR (500 MHz, CD₃CN, 70 °C) δ_H 7.38–7.22 (m, 10H), 7.04–6.95 (m, 1H), 5.85–5.60 (br b, 1H), 5.06 (d, J = 12.7 Hz, 1H), 5.01 (d, J = 12.7 Hz, 1H), 4.54–4.45 (m, 1H), 4.43–4.33 (m, 2H), 4.27–4.19 (m, 1H), 3.21–3.12 (m, 1H), 2.92 (dd, J = 14.1, 8.6 Hz, 1H), 2.54–2.43 (m, 1H), 2.25–2.14 (m, 1H). ¹³C RMN (125.7 MHz, CD₃CN, 70 °C) δ_C 176.0 (C), 172.7 (C), 157.2 (C), 138.7 (C), 138.5 (C), 130.7 (2 × CH), 129.74 (2 × CH), 129.68 (2 × CH), 129.2 (CH), 128.9 (2 × CH), 128.0 (CH), 67.6 (CH₂), 66.9 (CH₂), 57.6 (CH), 49.9 (CH), 39.2 (CH₂), 29.7 (CH₂). HRMS (ESI) calculated for C₂₁H₂₂N₂O₅Na [M + Na]⁺ 405.1426, found 405.1426. Anal. Calcd for C₂₁H₂₂N₂O₅: C, 65.96; H, 5.80; N, 7.33. Found: C, 66.22; H, 6.05; N, 7.45.

(2S)-N-[tert-(Butoxycarbonyl)homoserine lactone (7n). Obtained from aldehyde 6n (53.0 mg, 0.17 mmol) according to method A for the preparation of homoserine lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (n-hexane/EtOAc, 70:30), yielding lactone 7n (24.0 mg, 0.12 mmol, 70%) as a crystalline solid: mp 102–104 °C (from n-hexane/EtOAc); [α]_D: −1 (c 0.30, CHCl₃). IR (CHCl₃) ν_max 3432, 1783, 1713, 1503, 1161 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 5.09 (s, 1H), 4.44 (t, J = 9.7 Hz, 1H), 4.39–4.30 (m, 1H), 4.24 (ddd, J = 11.4, 9.3, 5.9 Hz, 1H), 2.80–2.70 (m, 1H), 2.25–2.13 (m, 1H), 1.45 (s, 9H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 175.5 (C), 155.6 (C), 80.7 (C), 65.9 (CH₂), 50.3 (CH), 30.8 (CH₂), 28.4 (3 × CH₃). HRMS (ESI) calculated for C₉H₁₅NO₄Na [M + Na]⁺ 224.0899, found 224.0897. Anal. Calcd for C₉H₁₅NO₄: C, 53.72; H, 7.51; N, 6.96. Found: C, 53.97; H, 7.45; N, 6.71.

(2S)-N-[Benzyloxycarbonyl)homoserine lactone (7o). Obtained from aldehyde 6o (67.4 mg, 0.20 mmol) according to method A for the preparation of homoserine lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (n-hexane/EtOAc, 60:40), yielding lactone 7o (14.3 mg, 0.06 mmol, 30%) as a crystalline solid whose characterization data have been reported (commercial compound) [86]. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 7.39–7.30 (m, 5H), 5.47–5.34 (s.a, 1H), 5.16–5.10 (s.a, 2H), 4.48–4.37 (m, 2H), 4.29–4.20 (m, 1H), 2.82–2.72 (m, 1H), 2.27–2.15 (m, 1H). HRMS (ESI) calculated for C₁₂H₁₃NO₄Na [M + Na]⁺ 258.0742, found 258.0743.

Methyl 2-oxo-1,3-oxazinane-4-carboxylate (7p). Obtained from aldehyde 6p (64.6 mg, 0.20 mmol) according to method A for the preparation of lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (n-hexane/EtOAc, 60:40), yielding lactone 7p (7.3 mg, 0.05 mmol, 23%) as an amorphous solid. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 5.34–5.21 (s, 1H), 4.45 (t, J = 8.4 Hz, 1H), 4.43–4.34 (m, 1H), 4.26 (ddd, J = 11.3, 9.3, 5.8 Hz, 1H), 3.71 (s, 3H), 2.82 ̶ 2.74 (m, 1H), 2.27–2.16 (m, 1H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 175.1 (C), 156.9 (C), 65.9 (CH₂), 52.8 (CH), 50.6 (CH₃), 30.6 (CH₂). HRMS (ESI) calculated for C₆H₉NO₄Na [M + Na]⁺ 182.0429, found 182.0430.

(2S)-N-Methyl-N-(p-toluenesulfonyl)homoserine lactone (7q). Obtained from aldehyde 6q (36.6 mg, 0.10 mmol) according to method B for the preparation of homoserine lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (n-hexane/EtOAc, 70:30), yielding lactone 7q (17.0 mg, 0.06 mmol, 62%) as a crystalline solid: mp 131–133 °C (from n-hexane/EtOAc); [α]_D: −33 (c 0.80, CHCl₃). IR (ATR) ν_max 1785, 1363, 1340, 1193, 1164 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 7.82–7.68 (m, 2H), 7.35–7.27 (m, 2H), 5.04–4.96 (m, 1H), 4.44–4.36 (m, 1H), 4.29–4.20 (m, 1H), 2.77–2.74 (m, 3H), 2.46–2.40 (m, 4H), 2.37–2.26 (m, 1H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 172.5 (C), 144.0 (C), 135.7 (CH), 129.8 (2 × CH), 127.7 (2 × CH), 65.3 (CH₂), 56.6 (CH), 30.3 (CH₃), 25.8 (CH₂), 21.7 (CH₃). HRMS (ESI) calculated for C₁₂H₁₅NO₄SNa [M + Na]⁺ 292.0619, found 292.0618.

N-Methyl-N-(p-fluorobenzoyl)homoserine lactone (7r) and (2S)-N-methoxymethyl-N-(p-fluorobenzoyl)homoserine lactone (7s). Obtained from aldehyde 6g (65.0 mg, 0.20 mmol) according to method B for the preparation of homoserine lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (n-hexane/EtOAc, 80:20), yielding the N-methylaminolactone 7r (14.7 mg, 0.06 mmol, 31%) as an amorphous solid and the lactone 7s (32.0 mg, 0.12 mmol, 60%) as a colorless oil.

Product 7r. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 7.55–7.48 (m, 2H), 7.11 (t, J_H,H = 8.6, J_H,F = 8.6 Hz, 2H), 5.27–5.03 (m, 1H), 4.61–4.49 (m, 1H), 4.40–4.28 (m, 1H), 3.02 (s, 3H), 2.65–2.40 (m, 2H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 173.6 (C), 171.5 (C), 163.7 (C, d, J_CF = 250.8 Hz), 130.9 (CH), 129.8 (2 × CH), 115.7 (2 × CH, d, J_CF = 19.3 Hz), 65.8 (CH₂), 55.4 (CH), 35.9 (CH₃), 25.6 (CH₂). HRMS (ESI) calculated for C₁₂H₁₂FNO₃Na [M + Na]⁺ 260.0693, found 260.0702.

Product 7s. [α]_D: −79 (c 0.80, CHCl₃). IR (CHCl₃) ν_max 1740, 1646, 604, 1420, 1055 cm⁻¹. ¹H NMR (500 MHz, CDCl₃, 26 °C) δ_H 7.51 (t, J_H,H = 6.9, J_H,F = 6.9 Hz, 2H), 7.11 (br t, J_H,H = 8.6, J_H,F = 8.6 Hz, 2H), 5.55–5.43 (m, 1H), 5.15–5.02 (m, 1H), 4.80–4.70 (m, 1H), 4.01 (dd, J = 12.3, 4.2 Hz, 1H), 3.82 (s, 3H), 3.64 (td, J = 12.2, 3.0 Hz, 1H), 2.33–2.10 (m, 2H). ¹³C RMN (125.7 MHz, CDCl₃, 26 °C) δ_C 171.1 (C), 169.8 (C), 164.0 (C, d, J_CF = 250.8 Hz), 130.1 (CH), 129.9 (2 × CH, d, J_CF = 4.6 Hz), 115.8 (2 × CH, d, J_CF = 22.0 Hz), 77.6 (CH₂), 65.4 (CH₂), 52.8 (CH₃), 51.0 (CH), 27.0 (CH₂). HRMS (ESI) calculated for C₁₃H₁₄FNO₄Na [M + Na]⁺ 290.0805, found 290.0805. Anal. Calcd for C₁₃H₁₄FNO₄: C, 58.42; H, 5.28; N, 5.24. Found: C, 58.19; H, 5.27; N, 5.11.

N-(Phenyloxycarbonyl)-N-(methyl)homoserine lactone (7t) and (2S)-N-(phenyloxycarbonyl)-N-(methoxymethyl)homoserine lactone (7u). Obtained from aldehyde 6p (64.6 mg; 0.20 mmol) according to method B for the preparation of homoserine lactones. After work-up and solvent evaporation, the residue was purified by radial chromatography (n-hexane/EtOAc, 70:30), yielding the N-methylaminolactone 7t (22.6 mg, 0.10 mmol, 48%) as a crystalline solid, and the lactone 7u (26.7 mg, 0.10 mmol, 50%) as a colorless oil.

Product 7t. Mp 72–74 °C (from n-hexane/EtOAc); [α]_D: −16 (c 0.61, CHCl₃). IR (CHCl₃) ν_max 3019, 1784, 1713, 1372, 1163 cm⁻¹. ¹H NMR (500 MHz, CD₃CN, 70 °C) δ_H 7.40 (br t, J = 8.2 Hz, 2H), 7.25 (br t, J = 7.9 Hz, 1H), 7.14 (d, J = 8.2 Hz, 2H), 4.85–4.71 (m, 1H), 4.43 (ddd, J = 9.0, 8.9, 3.1 Hz, 1H), 4.27 (dt, J = 7.1, 9.1 Hz, 1H), 3.06 (s, 3H), 2.62–2.45 (m, 2H). ¹³C RMN (125.7 MHz, CD₃CN, 70 °C) δ_C 175.2 (C), 152.9 (2 x C), 130.5 (2 × CH), 126.7 (CH), 122.9 (2 × CH), 66.8 (CH₂), 58.4 (CH), 34.3 (CH₃), 26.6 (CH₂). HRMS (ESI) calculated for C₁₂H₁₃NO₄Na [M + Na]⁺ 258.0742, found 258.0744. Anal. Calcd for C₁₂H₁₃NO₄: C, 61.27; H, 5.57; N, 5.95. Found: C, 60.92; H, 5.96; N, 6.05.

Product 7u. [α]_D: −96 (c 0.71, CHCl₃). IR (CHCl₃) ν_max 3019, 1724, 1438, 1415, 1065 cm⁻¹. ¹H NMR (500 MHz, CD₃CN, 70 °C) δ_H 7.45–7.35 (m, 2H), 7.29–7.21 (m, 1H), 7.17–7.09 (m, 2H), 5.62–5.38 (m, 1H), 5.10–4.95 (m, 1H), 4.75–4.60 (m, 1H), 4.01–3.92 (m, 1H), 3.83–3.77 (m, 3H), 3.66–3.56 (m, 1H), 2.30–2.17 (m, 1H), 2.15–2.06 (m, 1H). ¹³C RMN (125.7 MHz, CD₃CN, 70 °C) δ_C 172.3 (C), 154.4 (C), 152.7 (C), 130.6 (2 × CH), 126.8 (CH), 122.9 (2 × CH), 76.0 (CH₂), 65.5 (CH₂), 54.5 (CH), 53.3 (CH₃), 27.5 (CH₂). HRMS (ESI) calculated for C₁₃H₁₅NO₅Na [M + Na]⁺ 288.0848, found 288.0844. Anal. Calcd for C₁₃H₁₅NO₅: C, 58.86; H, 5.70; N, 5.28. Found: C, 58.72; H, 5.89; N, 5.16.

3.2. Biological Screenings

3.2.1. Quorum-Quenching Activity: Quantification of Violacein Production

The determination of the influence of the compounds on the production of violacein was carried out with the indicator strain C. violaceum CECT 494, according to the procedure reported by Choo et al. [60].

A standard 40 mM solution of the compounds (in DMSO) was diluted in LB medium so that after mixing 1 mL of the diluted solution and 1 mL of the inoculum, a final concentration of 200 μM was obtained (for the most active compounds, concentrations of 100 and 50 μM were also tested). In a similar way, the inoculum was prepared by diluting a preinoculum in LB media, so that after mixing with the product solution, the final inoculum density was 0.7–1 × 10⁸ CFU/mL. A control without treatment (vehicle only) was also prepared to compare the treated and untreated cultures.

The cultures were incubated at 30 °C, with constant shaking for 24 h. For the extraction of the violacein pigment, 1 mL of each culture was subjected to two cycles of centrifugation (Spectrafuge 24D Labnet centrifuge) at 14,000× g for 10 min. The first centrifugation cycle allowed removal of the medium (supernatant). The pigment was solubilized by cell treatment with DMSO, and in the second centrifugation cycle, the supernatant containing the dye was separated from the bacteria. Finally, an aliquot of the supernatant (200 μL) was added to 96-well plates, and the absorbance was read at 595 nm in the FLUOstar Omega plate reader, BMG LABTECH (Ortenberg, Germany). The products were tested in triplicate.

3.2.2. Antimicrobial Activity

The susceptibility of Gram-negative Salmonella enterica CECT 456, Campylobacter jejuni CECT 9112, and Pseudomonas aeruginosa CECT 108 and Gram-positive Staphylococcus aureus CECT 794 to the different compounds and concentrations was evaluated either with the broth microdilution method (for the larger lactone libraries) or the disk diffusion method (aldehyde precursors).

Broth microdilution method. The procedure was carried out in 96-well plates and followed the EUCAST recommendations for each microbial strain [59,60,61]. Two or three colonies of a microorganism were selected and incubated in liquid medium (MH broth or Sabouraud dextrose) with shaking for 18–24 h at 37 °C. The preinoculum was then subjected to serial dilutions, so that the concentration of the final inoculum was the recommended by EUCAST [59,60,61].

Meanwhile, the compounds were dissolved in DMSO to achieve standard (40 mM) concentrations. The standard was diluted in the liquid medium, so that after taking a 50 μL aliquot and mixing it with the same volume of the inoculum (50 μL) in the plate well, the final product concentration was 200 μM. The inoculated plates were incubated as commented for the preinoculum. Then, the absorbance was measured at 595 nm in the plate reader (FLUOstar Omega, BMG LABTECH). The wells where growth (or turbidity) was not visually observed were subjected to a viable cells count on agar plates. The products were tested in triplicate.

Disk diffusion method. To prepare the inoculum, the desired microbial strain was cultured on a Mueller Hinton (MH), LB, or Sabouraud 4% glucose agar plate and incubated for 18–24 h at 37 °C (exhaustion of media procedure). Then, 2–3 colonies isolated from this plate were introduced in a tube containing 3 mL of sterile physiological saline solution until a turbidity of 0.5 MacFarland was reached (measured with a Grant biodensitometer DEN-1B), which corresponded to 1–2 × 10⁸ CFU/mL.

Meanwhile, paper disks containing the potential antimicrobial were prepared. The disks were impregnated with standard (12.5 mM) solutions of the compounds in 7:3 ethanol/DMSO mixtures so that each disk contained 0.25 μmol of the potential antimicrobial. For the positive controls, the disks were impregnated with tetracycline (30 μg, 0.06 μmol). For the negative control, the disks were soaked into the vehicle.

Once the inoculum and the disks were ready, MH-agar plates were inoculated (100 μL of the inoculum), and then the disks were placed on the top. The plates were incubated for 18–24 h at 37 °C, and afterwards the inhibition zones were measured in mm. The products were tested in triplicate.

3.3. In Silico ADME Studies

The in silico ADME studies were performed with the SwissADME tool, developed by the Swiss Institute of Bioinformatics [66,67] as commented on in the text and the references.

4. Conclusions

In summary, a library of AHLs with a variety of N-substituents and a library of AHL aldehyde precursors were prepared in good yields and from readily available, low-cost hydroxyproline substrates. An initial oxidative radical fragmentation, which cleaved the pyrrolidine ring, afforded unusual N-substituted 4-oxohomoalanine derivatives in good yields as pure enantiomers. Then, a one-pot reduction–lactonization reaction under two different conditions gave a variety of AHL derivatives, including N-substituted amino lactones with N-acyl, N-carbamoyl and N-sulfonyl groups or N,N-disubstituted amino lactones, in high optical purity.

In order to identify a potential antibiotic, the antimicrobial and quorum-quenching activities of the library were evaluated. To determine the quorum-quenching activity of lactones, the reporter strain of the Gram-negative pathogen Chromobacterium violaceum CECT 494 was used, and the generation of the violet pigment violacein was measured. For the first time, sulfonamides and benzamides of related AHLs were compared. Also, for the first time, the activities of N,N-disubstituted AHLs were compared with those of their N-substituted analogs.

Sulfonamides 1a–c and 1e, the benzyl carbamate 1o, and the N-methyl toluenesulfonamide 1q were the most active compounds. In contrast, the benzamides 1f-m displayed little quorum-quenching activity. To our satisfaction, the most active QQ lactones presented low antimicrobial activity against C. violaceum CECT 494, a requisite for pure QQ agents. The activity against S. aureus CECT 794, C. jejuni CECT 9112, Salmonella enterica CECT 456, and P. aeruginosa CECT 108,was also evaluated, observing low activity.

In contrast, some of the aldehyde precursors displayed antimicrobial activity. The sulfonamide derivatives 2b–e and the dinitrobenzamide 2k displayed promising activity against S. aureus CECT 794 and C. jejuni CECT 9112, compared with the respective lactones. It suggests an important role of the 4-carbonyl group in the interaction with biological receptors.

Finally, in silico ADME studies carried out with the SwissADME tools suggest that these compounds have low toxicity and favorable ADME properties.